1
PROMOTING SELF DIRECTED LEARNING SKILLS
IN FIRST YEAR STUDENTS
ABSTRACT To be successful university learners, first year students need to develop skills in self directed
learning. This encompasses a range of cognitive and meta-cognitive skills including generating
one’s own learning goals, planning how to tackle a problem, evaluating whether learning goals have
been met, and re-planning based on this evaluation. Based on qualitative analysis of online learning
logs written by 113 first and second year computer science students about a creative project, this
paper describes in detail the strategies used by these students in terms of planning and managing a
project, generating and selecting creative ideas, and monitoring and evaluating their own progress.
A summary of the behaviour shown by effective learners and less effective learners in each of these
areas is given, as well as suggestions of prompts which educators could use to structure a learning
environment intended to develop self directed learning skills.
INTRODUCTION There is increasing recognition of the importance of self directed learning within higher education.
As well as learning domain specific facts and concepts, students are expected to take responsibility
for managing their own learning, requiring them to learn a range of transferable cognitive and
meta-cognitive skills which will enable them to learn successfully in the future. The Quality
Assurance Agency for Higher Education, the body responsible for checking that universities
maintain academic standards in the UK, recently conducted a project in Scotland relating to
“Enhancing the First Year Experience” (Mayes, 2009). The project recognised that the learning
habits acquired by students in their first year were crucial with respect to their success as learners
in future years of university and beyond. A key theme of the work was that of empowerment: “The
question here is how to equip students, as close to the start of their studies as possible, with the
skills, capacities and knowledge to be effective as independent learners for the rest of their
programme, and for their subsequent employability, professional development, and, for that
matter, lifelong learning”(Mayes, 2009;4). The term empowerment is also referred to as “self
regulated learning” in the report (p12) and includes the skills of self assessment and reflection. The
related concept of “learning how to learn” is predicted by the Beyond Current Horizons Project to
be an important driver in education over the next thirty years: “It is, however, essential that
learning to learn does become a key feature of the future of education, to ensure that at the heart of
education is learning to be human and to take responsibility for one‘s place in a society which
encourages and enables participation by all its citizens, to enable them to fulfil their own potential
and shape the future for subsequent generations” (Higgins, 2009;13). Higgins notes that “Learning
how to learn” is a broad and sometimes nebulous concept which encompasses independence, meta-
cognition, thinking skills, self regulation, self esteem and self efficacy.
2
Given the importance attached to the development of these skills, how can university educators
help their students develop them, particularly in the crucial first year? This paper attempts to
answer this question based on qualitative analysis of learning logs written by 113 students on a
computer science introductory programming module. By examining the reflective strategies of the
most capable self directed learners, in contrast to the writing of learners who did not engage so
productively in written reflection, it is possible to identify approaches which educators could use in
the future to prompt effective self directed learning behaviour. As the students were writing about
their experiences of working on an open ended creative project, it is anticipated that the current
findings will be relevant to students and educators in disciplines other than computer science
which require skills in conducting independent creative projects.
The paper begins with an overview of the design of the learning environment under study and uses
this to contextualize the literature review on self-directed learning and the software tools which
may promote it. In order to increase conceptual clarity, it reviews and attempts to disambiguate
commonly used terms in the literature: self directed learning, self regulated learning and meta-
cognition. It also considers previous research into software tools and learning activities designed to
promote meta-cognitive skills and discusses the degree to which such environments should
attempt to explicitly structure student reflection. An analysis of the corpus of student learning logs
follows. By analysing the effective and less effective strategies used by students in their learning
logs, the paper offers advice to other educators which may assist them in designing a learning
environment in which self directed learning skills can flourish.
OVERVIEW OF LEARNING CONTEXT Before considering previous research in the area of self-directed learning, it is useful to describe the
context in which the present study took place in order to meaningfully relate the design of this
learning environment to designs proposed in previous research. The introductory module in
Interactive Systems was developed as part of newly restructured degree programmes in Computer
Science and Information Systems at a small engineering focussed university in Scotland. The
objectives of the module were to give students a motivating introduction to programming and
multimedia systems development lifecycles and to help them to develop study skills which would
be useful throughout their university careers. This compulsory module ran in the first semester of
first year, for 12 weeks. Students attended one lecture and one two hour lab per week. The module
was assessed by 100% coursework in the form of a portfolio comprised of a learning log (25%),
peer reviews (25%) and an interactive pet created in the Second Life programming environment
(50%). The interactive pet marks were allocated for both the visual appearance of the pet which
required 3D object building skills, and its interactive behaviour which required programming skills.
There were several elements in the pedagogical design of the module. One of the most important
aspects was that the students should work independently on a creative project within a supportive
environment. It was intended that the students should develop skills in directing their own
projects, from choosing a suitable project (in this case deciding which pet to make) to evaluating
the success of their enterprise. Another important aspect was to encourage the students to work
effectively with their peers. There were various opportunities for peer learning: working in
3
problem solving pairs in lectures, pair programming in labs, formal peer review of pets in week 10,
and informal support in the form of peer blog comments throughout the semester. A third design
element for the module was that it should be a positive affective experience for the students. As
learning programming can be a frustrating and demoralising task, we aimed to counterbalance this
by providing a fun, supportive and encouraging atmosphere in which students could build
confidence and take pride in their work. An example of this was celebrating student achievements
with “Creature of the Week” where the lecturer displayed a screenshot of a student’s pet in
progress and explained what was good about it.
The Second Life environment was considered to be appropriate for a first year programming course
because it both enables novices to create 3D objects and to make them interactive using the Linden
Script programming language. An interesting pedagogical affordance offered by Second Life is that
the students’ ongoing work can be left on display for all other users of the learning environment to
see. Students can also see each other at work building new objects and discuss their work with
others. This is unusual in computer science classes, because works in progress do not normally
have a visual component which can be shared with peers or instructors. We used a private island to
which only students registered for the module had access to give them privacy until the end of the
module when we invited some colleagues to visit an exhibition (“Virtual Crufts”) of their interactive
pets. An example of a student’s pet is shown in Figure 1. The dragon has the interactive behaviours
of breathing fire, laying an egg which hatches into a baby dragon, and eating a squirrel which is
summoned by using an acorn as bate.
4
Figure 1.A dragon pet created by a first year student.
Although the module was designed for first year students, an unexpected temporary difficulty with
the restructuring of the degree programmes made it necessary for second year students to take the
module too for this year only. This made teaching considerably more difficult because it widened
the range of prior experience and background knowledge of the learners, but it did have the
advantage of providing data on the self directed learning strategies of more experienced students
which proved very helpful for analysis.
LEARNING LOGS The focus for this paper is on analysis of the students’ self directed learning skills as demonstrated
in their learning logs. The instructions for completing the learning log in the assignment
specification are shown in Figure 2.The section of the marking scheme relating to learning logs is
shown in Table 1. This marking scheme is similar to others used within the department, and was
issued to the class at the start of semester. At the second lecture the instructors explained in detail
what was required in a learning log, what sorts of things the students should write about, gave
some examples of good reflective writing and demonstrated how to use the blogging software in
the Blackboard virtual learning environment which is used throughout the university. Students
were divided into groups of five and asked to comment regularly on their group members’ blogs.
5
Students could also read and comment on the blogs of any other classmates, but the blogs were not
available to users outside the class.
Figure 2. Learning log instructions
Mark range Marking indicator (%) <40 Unacceptable. There is no reflection on learning.
40 -49 Limited discussion of design problems encountered, or solutions used to address them. Irregular, infrequent entries.
50 – 59 Adequate discussion of design problems encountered. Less on how problems were solved. Regular but less frequent entries (weekly).
60 – 69 Good discussion of design problems encountered and strategies used to solve them. Regular entries (after every class, for example).
70 > Excellent discussion of design problems encountered and strategies used to solve them. Thoughtful reflection on how ideas have changed and effectiveness of design process. Regular entries (every couple of days or more).
Table 1. Marking scheme for learning logs.
The students were explicitly prompted to write blog entries at the end of weekly lab worksheets.
Lab sheets in the early stages of the module contained prompts such as “Reflection. 15 minutes
before the end of the session, stop working in SL and think about what you have done in this lab.
Write in your blog about:
a. What you have learned today
b. What problems did you come across and how did you solve them?
c. What do you want to work on next time?”.
As the module progressed, the lab worksheets contained less specific prompts of what sort of
questions to consider, giving only more general reminders such as “Remember to write about what
you have learned in your blog, and to comment on your group members' learning logs”.
“Throughout the project you will keep a learning log of what you have learned. A learning log is
a diary of what you learned throughout the module. This will help us to assess the rest of your
work – it is your chance to demonstrate the learning process you went through to arrive at your
portfolio. This means that even if you encounter lots of problems and don’t achieve what you
were aiming for, you can still get credit for trying. The learning log should include a record of
your ideas and how they changed as you learned more, and a log of the problems you
encountered and how you fixed them. The last entry should reflect on how successful you think
your portfolio is and why. You are expected to update the blog regularly throughout the term.
You will keep your learning log using the blog facility in Blackboard. You are expected to read
blogs written by the other members of your group, and give them helpful comments. In return,
their comments will help you. We will also read your blog from time to time to give you
feedback.”
6
DEFINITIONS OF SELF-DIRECTED LEARNING There is a lack of clarity in the literature about the definitions of meta-cognition, self-regulation and
self-directed learning (Kaplan, 2008, Alexander, 2008, Loyens et al., 2008).Starting from the
concept of self directed learning, Figure 3 illustrates a hierarchy of component skills derived from
definitions of concepts proposed by other researchers. It can be viewed as a map of the component
skills of self directed learning, its relationship to affect, and its impact on performance. It was used a
guide to the sorts of behaviour which might be documented in the students’ learning logs during
the analysis described in the Qualitative analysis section.
The hierarchy is intended to be helpful in showing the relationship between different terms, and
therefore to aid conceptual clarity. It has the advantage of highlighting inconsistencies or
disagreements between previous researchers. In particular, Kaplan (Kaplan, 2008)suggests that
meta-cognition and self-regulated learning, although related, are distinct from each other, and in
fact should be considered as subtypes of a more general capacity of self-regulated action. In view of
the consensus of definitions of the concepts proposed by other researchers, however, it seems more
helpful to think of a hierarchy of concepts in which meta-cognition is a component skill of self
regulated learning (i.e. a parent/ child instead of a sibling relationship).
Figure 3. A hierarchy of self directed learning concepts.
7
Self directed learning is a process in which the learner takes the initiative and responsibility for
setting his own learning goals, identifying and addressing gaps in his learning, identifying
resources, selecting and carrying out learning strategies and evaluating his own learning (Loyens et
al., 2008) The learner may receive varying degrees of teacher or peer support for these tasks.
(Loyens et al., 2008)compare self-directed learning with self regulated learning in the
context of problem based learning, identifying that self directed learning is a broader concept which
includes self regulated learning. In a self directed learning environment, students have more
freedom to generate and pursue their own goals, and undertake critical evaluation of the materials
they select. The self directed learner initiates the learning task, whereas in self regulated learning,
the task could be set by the teacher. In the present study, the learners were directed by the teacher
in the sense that they were asked to produce an interactive product, but the choice and design of
that product were not specified. This is perhaps similar to a teacher setting the problem to be
studied in a problem based learning course, which is considered by Loyens to be self directed
learning: the problem is specified, but the method of solution or the solution itself is not. Therefore
the assignment undertaken in the module in the present study can be considered to be on the
spectrum of self directed learning activities. As is often the case in problem based learning
activities, the students on this module were asked to start working on their task before they
received formal curriculum inputs on it, such as lectures or text book material. This required the
students to identify deficits in their knowledge, and make plans to address them.
In self regulated learning environments, learners have control over directing their cognitive
and meta-cognitive resources to achieve a learning goal. Actively engaging in the process of their
own learning, they set themselves sub goals, plan their activities, evaluate the success of those
activities and replan their work based on this. Self regulated learning involves meta-cognition,
intrinsic motivation and other affective factors, such as willingness to make an effort and
persistence (Vrugt and Oort 2008). (Vrugt and Oort 2008)developed a model of effective self
regulated learning derived from path analysis of exam scores and self report questionnaire data of
952 first year learners. They found a positive relationship between intellectual ability and self-
regulation learning, and confirmed that more active engagement in learning and self regulation
resulted in higher exam scores. Students who took a mastery approach to learning (i.e. those who
believed that effort leads to an improvement in performance) were better able to deal with
resource management, and were more inclined to employ meta-cognition. In contrast, less effective
self regulators tended to focus on avoiding negative performance (e.g. trying to avoid appearing
stupid) which was not linked to the use of meta-cognition, or deep cognitive processes. There is
some evidence from this study that a combination of mastery and performance approach goals can
be effective; intrinsically motivated students can also benefit from the extrinsic motivation of
avoiding being judged negatively for poor work.
(Coutinho and Neuman 2008) found that self efficacy – a person’s perception of their own ability
to achieve success- was a strong predictor of performance among undergraduate psychology
students, whereas meta-cognition was a weaker predictor. The use of mastery and performance
approach goals were predictors of self efficacy; the students who focussed on developing their
competencies and seeking knowledge and understanding of an area, and those who aimed to
8
demonstrate competency to others tended to believe in their ability to succeed more than those
who tried to avoid mastery, or focussed on avoiding showing a negative performance to others. This
is consistent with the findings of (Vrugt and Oort, 2008). Based on these findings, Coutinho and
Nueman recommend that teachers should aim to foster self efficacy and observe that “the richest
learning environment could be one that fosters support and encouragement and helps to build faith
in abilities and skills” (Coutinho and Neuman, 2008;147). In the domain of Information Systems,
Moores, Chang and Smith report that self efficacy was a predictor of declarative and procedural
knowledge among Management Information Systems students who answered MCQs about a
diagram formalism, and constructed diagrams based on it (Moores et al., 2006). They found that
meta-cognition was related to procedural knowledge (i.e. constructing the diagram) but not
declarative knowledge (answering the MCQs).
Self efficacy and meta-cognition, then, both appear to be important components of self directed
learning. Interestingly, while meta-cognitive abilities tend to develop with age, self efficacy usually
drops during formal education(Noushad, 2008).
(Flavell, 1979)believed that meta-cognition consisted of interactions between meta-cognitive
knowledge, meta-cognitive experiences, goals and strategies. He considered meta-cognitive
knowledge to be world knowledge about the cognition of oneself and other people, and the nature
of tasks. Meta-cognitive experience refers to cognitive and affective experiences arising from and
relating to intellectual tasks. These can arise at any stage of a task, and can be of varying length
from a fleeting sense of puzzlement to an ongoing feeling that one is likely to fail. Flavell believed
that such experiences were likely to arise during tasks which stimulate “careful, affect highly
conscious thinking” (Flavell, 1979;908)because “such situations provide many opportunities for
thoughts and feelings about your own thinking to arise, and in many cases, call for the kind of
quality control that meta-cognitive experiences can help supply”(Flavell, 1979;908). He also
acknowledged that meta-cognitive knowledge and experiences have some overlap. Meta-cognitive
goals are the “objectives of cognitive enterprise” and meta-cognitive strategies are used to regulate
the progress of cognitive tasks. Examples of meta-cognitive activities include planning, predicting,
guessing, and monitoring (Noushad, 2008). An aspect of meta-cognitive monitoring is self
appraisal: an activity which requires critical evaluation of one’s own performance in order to take
steps to correct it. Noushad notes that learners require a supportive affective environment in order
to feel comfortable with the challenging task of self appraisal.
However, Brown suggests that the term meta-cognition is a term that could be “pensioned off”
(Brown, 1987;106)for the sake of clarity in the field. She points out that confusion arises because it
refers to both knowledge about cognition and the regulation of cognition. In her review of meta-
cognition research, she discusses self regulation to be an aspect of meta-cognition, which is the
inverse of the relationship suggested by Vrugt and Oort. However, she does suggest that to increase
clarity, the term “meta-cognition” could be used to refer to knowledge about cognition only,
whereas “process terms” such as planning ahead, monitoring, resource allocation, self questioning
self- directing etc could be used alone without the addendum, meta-cognition” (Brown
1987;106)This view is reflected in the diagram, where the process terms have been included as
component skills of self regulation.
9
A highly related concept to self directed learning is affect. Boekaerts in considering the
relationship between motivation, affect and learning, states that “emotions experienced during goal
pursuit are sources of information that students interpret and use” (Boekaerts, 2003;186)This is
consistent with Flavell’s position on meta-cognitive experience, particularly because of Boekaert
and colleagues’ empirical evidence that students’ evaluation of their own performance is influenced
by their emotional state. Students experiencing a higher intensity of positive emotion were likely to
assess their performance more favourably and vice versa. Similarly, students who believed they had
the resources to complete a task, had confidence in their own ability to complete the task and who
enjoyed the task were more likely to evaluate their performance as successful afterwards. They also
observed that students tended to experience negative emotions when they realised that there were
obstacles in the way of achieving their goal, or if they realised their progress was slow. This
illustrates that the meta-cognitive behaviours of goal setting and monitoring are mediated by the
learner’s emotional state; affect and meta-cognition are intertwined. This was apparent from the
analysis of the learning log corpus described in the Qualitative analysis section.
REFLECTIVE WRITING ONLINE Over the last decade, there has been interest among university educators about how online tools
such as blogs can be co-opted to support learning. There appears to be a natural fit between current
goals of higher education and the affordances of blogs in two main areas. Firstly, blogs can be seen
as a convenient way to produce and share a reflective learning log and secondly, they can offer an
audience for students writing within the safety of a learning community thus offering opportunities
for collaborative learning.
(Boud, 2001;356) considers how reflective writing may promote learning. Reflection is the process
of exploring an experience in order to learn something new from it. There is an important
emotional component involved; Boud writes “Reflection involves taking the unprocessed, raw
material of experience and engaging with it as a way to make sense of what has occurred. It
involves exploring often messy and confused events and focusing on the thoughts and emotions
that accompany them.” (Boud, 2001;10). This is consistent with Boekart’s emphasis on the
interweaving of affect with cognition and metacognition (Boekaerts, 2003). A written journal can be
a place both for both recording learning experiences and processing them in order to understand
their implications. Occasions which can be appropriate for reflective journal writing include writing
in anticipation of events, reflection in the midst of events, and reflecting after undergoing a new
experience. In the context of a learning log within the Interactive Systems module, the students
could usefully undertake each of these activities in planning their portfolio work, looking back on
lab sessions or writing about current unsolved programming problems, and in looking back over
the whole semester to evaluate the trajectory of their learning experiences throughout the module.
Boud also points out some barriers to reflective writing in a learning situation. There is the issue of
privacy: to what extent can students feel comfortable writing about the emotional component of
their feelings honestly when they know it will be read by their peers and staff? This is compounded
by the fact that learning journals are often assessed (as they are in the Interactive Systems module).
Students may be reluctant to write about misconceptions or difficulties they have with the course
10
material in case this results in a negative assessment of their abilities. It is therefore important to
clearly explain to students who their audience will be, and how the journal will be marked. In the
Interactive Systems module we emphasised that rather than deducting marks for accounts of
problem solving, such writing would help us to understand their learning process and so give them
credit for the strategies they tried even if ultimately they did not find a solution to the problem.
With the increase in popularity of blogging software in the early 2000s, educators began to see the
potential of it for learning situations including reflective writing. In 2004, Williams and Jacobs
concluded that “blogs have the potential, at least, to be a truly transformational technology in that
they provide students with a high level of autonomy while simultaneously providing opportunity
for greater interaction with peers”(Williams and Jacobs, 2004;245). In a survey of 104 MBA
students' attitudes to blogging, they found that 77% of students believed that writing and
commenting on blogs enhanced student interactivity and increased “the level of meaningful
intellectual exchange between students”. The authors recommended that it is important to give
students direction about how to blog and what benefits they might expect to experience from it.
(Ferdig and Trammell, 2004) commented on the potential educational benefits of blogging in an
educational setting, remarking that they are ideal spaces for students to make meaning and publish
their reflections, thoughts and understandings for an authentic audience. They identified the
comments feature as useful for promoting peer feedback and scaffolding of new ideas. Additionally,
in their view, blogging can increase student ownership and interest, give opportunities for students
to legitimately interact within a community of practice and celebrate diverse points of views.
However in spite of this potential, more recently (Kerawalla et al., 2009) warn that it can be difficult
for educators to implement blogging as a successful learning activity, citing a range of previous
studies which reported lack of engagement in blogging by students. In response to this, the authors
developed a framework for educators and learner bloggers based on interviews with Masters levels
students from an e-learning course who had written reflective blogs on a voluntary basis during
their course. The students used the blogs for a variety of purposes including carrying out course
related activities to share with other students, carrying out course activities but for their own
benefit rather than for sharing, keeping a learning journal and keeping a personal online document
store. In their framework the authors propose a set of questions designed to encourage students to
think about how to begin and subsequently sustain a blog. The questions are posed at a high level,
referring to the audience, community, and presentation of the blog as well as technological
considerations. These questions are designed to be used when designing a blogging activity or
beginning a blog, rather than to prompt specific sorts of reflection on learning during a blogging
session.
Luehmann discusses the use of blogging as a means of reflection in adult learners by examining a
detailed case study of a science teacher who blogged her experiences of teaching in an inner city
school(Luehmann, 2008). Luehmann argues that the teacher used the blog as a tool to develop her
professional identify, and to think critically at a distance about her classroom practice and more
general issues relating to education in her local context. The teacher found the blog useful as a
method for recording, easily retrieving and revisiting her own thoughts and experiences over time.
The author concludes that the blog format offered the teacher an opportunity for self directed
11
learning and meta-cognition by enabling her to write about issues which were important to her at
the moment they cropped up, thus providing the benefits of a personal learning journal combined
with the motivational power of a real audience. The researcher concludes “This study also provides
empirical support for some of the claims made in the blogging literature about the unique learning
affordances blogging can offer learners in general and teachers more specifically. These include the
opportunity to be self-directed, to reflect and engage in meta-cognition, to further develop one’s
thinking because of interaction with an audience, to play the role of knowledge-broker, and to
engage in identity work.” (p335). Luehmann's approach was to analyse the blogging of one
exceptionally committed blogger in depth with a view to understanding the educational benefits
which can emerge from blogging in the best case. She then identifies ways in which these benefits
could be harnessed for other learners in the future, perhaps in the forms of assignments for trainee
teachers or those in a CPD setting. A similar approach is taken in the present study: by examining
the writing of highly capable self directed learners, the paper identifies the sorts of blogging
activities which may be beneficial for students who are less effective self directed learners.
Although reflective writing skills have not traditionally been part of computer science curricula
(Stone and Madigan, 2007) highlight the importance of developing effective written communication
skills in undergraduate computer science and information systems students. After describing a pilot
project in which students gradually developed skills in reflective writing, the authors conclude that
“A detailed approach that is able to show a gradual progression in metacognitive development is
needed to fully reap the benefits of reflection and experiential learning.”(Stone and Madigan,
2007;45). This paper attempts to do this by benchmarking the skills displayed through reflective
writing by first and second year students on these degree programs.
From the perspective of the benefits of blogging on collaborative learning, Minocha provides a
literature review of educational uses of social software, including blogs. “Social software seems to
match well with modern thinking about educational practice. In particular, it promises learners of
new opportunities to be independent in their study and research. Social software tools encourage a
wider range of expressive capability. They facilitate more collaborative ways of working and they
furnish a setting for learner achievements to attract an authentic audience”(Minocha, 2009b;356).
She identifies how social software is consistent with educational theories such as collaborative
constructivism, and documents that common applications of blogs in higher education include
active learning, reflective practice and learning journals.
In a companion publication, the author reports on empirical data on the use of social software in
higher and further education, based on 26 case studies from UK institutions (Minocha, 2009a). The
findings enumerated benefits to educators including being able to track students' progress and give
feedback early, and being able to teach more interactively. They also identified challenges to the
educator of such software tools, such as the lack of guidelines for designing effective learning
activities using such software or the workload involved in providing feedback on high volumes of
student writing, particularly for student blogs. Benefits for students included the development of
collaborative skills, for example when constructing group reflective blogs and giving and receiving
peer feedback through blog comments. Challenges for students included worries about their work
being in the public domain (although student work can be protected in a university VLE as in the
12
study reported here), the learning curve of getting to grips with new technology and deriving
educational value from using the tools in situations where their co-learners did not contribute, for
example by commenting on blog posts.
TOOLS FOR SUPPORTING SELF DIRECTED LEARNING While some higher education researchers have focused on the educational benefits which may
emerge from adapting pre-existing software such as blogs to learning activities, there is a long
tradition within the learning sciences community of designing new software for specific learning
interventions in order to test theories of learning. There have been several successful attempts at
creating learning environments to scaffold learners’ self directed learning skills, particularly in the
domain of scientific inquiry and among younger learners. Findings from empirical classroom
studies have often been incorporated into the design of software tools to support learners in meta-
cognitive and reflective tasks.
A key insight used in previous research is that self directed learning skills can be enhanced by
making the learner explicitly aware of the process of learning. (Collins et al., 1991)describe the
notion of cognitive apprenticeship in which the normally invisible process of thinking is made
visible to both teachers and learners. They summarise a set of principles for designing learning
environments in which cognitive apprenticeship might thrive. Their methods for developing
expertise include the teacher modelling, coaching, and then scaffolding a task for learners followed
by encouraging students to articulate their thinking, reflect on their performance and explore a
topic by finding their own problems to solve. These methods were applied during the Interactive
Systems module: the staff modelled how to approach writing programs during lectures, and
coached the students as they attempted to solve problems designed by the staff to scaffold the
further development of programming skills. It was, however, equally important to help students to
develop problem setting skills; in this case by asking them to propose an interactive pet of their
choice which they would develop for their portfolio work. Reflective writing in blogs was an
opportunity for learners to reflect in an open ended fashion on their progress, thus serving as an
free-form external representation of their self-directed learning skills.
(Scardamalia and Bereiter, 1991)developed the landmark Computer Supported Intentional
Learning Environment (CSILE) in order to give higher level cognitive and meta-cognitive processes
an external representation through which learners could reflect on and reason about these
normally hidden skills. Part of their agenda in developing this software was to promote a particular
model of teaching in which the teacher aims to scaffold learners to “formulate their own goals, do
their own activating of prior knowledge, ask their own questions, direct their own inquiry and do
their own monitoring of comprehension” (Scardamalia and Bereiter, 1991;39) In this model, the
teacher is attempting to help the learner grow out of dependency on the teacher and therefore
assume increasing responsibility for their own learning and becoming self directed learners. This
goal is of particular relevance in the education of first year university students who will require this
sort of autonomy to be successful in the later years of their studies and beyond. An interesting
design constraint on such a software environment is that while in-built scaffolding can facilitate
users in developing effective plans, ideas and questions, its very purpose is to encourage learners
13
towards autonomy and away from externally provided guidance. Care must be taken to strike a
balance between enabling and restrictive structural templates in the software. The authors urge
caution when designing in-built procedural prompts until the designers clearly understand the
stages in the learning process at which learners typically become blocked from achieving their
goals. In such cases, carefully designed sentence opener prompts can be beneficial in that they
encourage novice learners to mimic the sorts of questions experts might ask themselves at certain
stages of the learning process. For example when planning an essay, a skilled writer might consider
questions such as “is this argument likely to be convincing to my reader?”. This is precisely the
approach taken in this paper: by examining the effective self-directed learning strategies used by
more able reflective writers in a relatively unstructured learning environment, it identifies prompts
which may be useful to learners throughout the stages of their creative projects.
(White and Frederiksen, 1998)also researched the use of structured prompts in a software tool
designed to promote self-directed learning. They developed and evaluated a computer based
curriculum (Thinking Tools Inquiry Curriculum) which was developed to support meta-cognitive
processes in learners in the domain of scientific inquiry. The software provides a series of prompts
which encourage appropriate types of thinking at various stages of the inquiry process.; as the
students iterate through the inquiry process, the software fades the initially heavy scaffolding until
eventually the students assume responsibility for their own reflection at each stage. For example,
the tool provides criteria for self and peer assessment of the quality of research under categories
such as “being inventive” or “being systematic”, both of which are important in a creative design
project such as the Interactive Systems portfolio. In the Thinking Tools curriculum, the teacher
initially models and coaches the use of these evaluation criteria, encouraging the learners to
develop autonomy in monitoring their own performance and progress and that of other learners. A
similar approach was used during the Interactive Systems module where staff gave examples of the
sort of features students should be considering when evaluating their peers' work.
White and Frederikson's findings indicated that the reflective self and peer assessment aspects of
the tool enhanced student learning, particularly among low-achieving students. Those students
who had mastered the self assessment criteria over an extended period of time produced higher
quality projects than their peers in the control condition. That is, reflective skills can improve with
practice. Furthermore, those students who took part in the reflective assessment condition were
more likely to submit completed projects; it would seem that the ability to reflect has a positive
impact on determination to succeed and persistence, which are both affective components of self-
directed learning.
More recently, Loh et al describe the Progress Portfolio (Loh et al., 1998) which was designed to
promote reflective inquiry skills in middle and high school students. In common with the previously
described systems, a design goal of this software was to make visible and explicit the normally
hidden process of inquiry, enabling students to reflect on their progress throughout an
investigation. Through representations of their progress on the interface, students are better
equipped to “organise, evaluate, and monitor the progress of an investigation” by keeping “inquiry
plans and candidate explanations in mind, systematically interpreting their collected data,
periodically reflecting, evaluating their progress and replanning.”(Loh et al., 1998;627). An
14
interesting aspect of the design was “templates” which could be pre-specified by teachers to
provide some structure around which students could base their reflections. These took the form of
prompts such as “How does this relate to your research question?” or a heading such as “Next
steps”. These templates were introduced after an initial field study in which students tended to
focus on action oriented activities at the expense of reflection. However, the researchers’ next field
study illustrated that while the templates were useful in supporting reflection, the students also
found them restrictive. For example, they could not include extra information in the templates
which the teacher had not anticipated. While this particular problem would be technically easy to
address, the authors also observe that “students are masters of efficiency – unless there is a
tangible benefit to a task, they won’t do it”(Loh et al., 1998;632). A general difficulty with teacher
specified prompts for reflection is that students will ignore them or complete them in a cursory
fashion unless they believe that they will find it useful later.
How best, then, to specify prompts which learners will find useful? (Davis, 2003;103)investigated
reflective prompts to aid middle school science students in a software environment called KIE
(Knowledge Integration Environment). She studied students’ responses to directed and generic
sentence starter prompts in order to investigate whether they simply needed reminders to stop and
reflect on their progress, or whether they would also benefit from advice on how to reflect
productively. An example of a directed prompt issued at the “Thinking back” stage of the project
was “”As we worked on this project, we wish we’d spent more time on…”; a generic prompt at the
same stage was “Our ideas right now are …” . Students’ answers to generic prompts have the
interesting property that they capture a baseline of their reflective capabilities, making their
thinking visible both to the learners themselves and teachers who are attempting to assist them. As
some students may lack the capability to reflect productively on their own, it seemed reasonable to
consider directed prompts in the light of “a more able other, prodding the students to consider
issues they would not have considered otherwise”(Davis, 2003;103)
However, findings indicated that in fact generic prompts resulted in better understandings of the
complex science projects than directed prompts which tended to result in unproductive reflection.
This somewhat unexpected effect was particularly pronounced for students who had a medium
level of autonomy according to a self-report questionnaire relating to their beliefs about learning. In
confirmation of the view that poor meta-cognition skills are a hindrance to cognitive performance,
poor reflectors in this study were less successful in completing the project task. Neither directed
nor generic prompts appeared to help the low autonomy students. The researchers discuss various
interpretations of their findings, including the notion that if students cannot understand specific
prompts, they may struggle and ignore them. By contrast, students who respond to generic prompts
are partly defining or interpreting the prompts for themselves and so what they write corresponds
to their own thinking and is more likely to be coherent. The author speculates that although prior
research has found that directed prompts are beneficial for learning about specific scientific
concepts, “a generic prompt may be more useful than at least some directed prompts when asking
students to engage in written reflection in large and complex projects with multiple instructional
goals…”(Davis, 2003;130).
15
It would appear from this work that generic prompts can have advantages over directed prompts,
although Davies acknowledges that they are unlikely to be effective on their own; they should be
used as an early warning system in which teachers can identify students who may be struggling
with effective reflection and can use this as the start as a learning conversation in which they try to
develop the necessary meta-cognitive sub-skill with the student in the context of their own project.
In fact, this is similar to the approach taken in the present study. The students used a flexible and
open ended tool for reflective writing (a blog), supported by increasingly general prompts on lab
worksheets and the assignment specification. The teaching staff attempted to respond to blog
entries using the blog comment feature where time permitted, and also gave advice and help about
project management during lab classes.
In summary, structured software support for self-directed learning skills, particularly reflective
writing can be effective in improving learning outcomes. However, care must be taken in designing
templates or prompts to guide learners to reflect in particular ways. The designer or educator
requires an in-depth understanding of the relevant learning process, such as information about the
sort of self-prompting experts might do when completing a similar task. The prompts should not be
too prescriptive in order to encourage student autonomy, and the benefits of writing responses to
the prompt should be apparent to the student for motivational reasons. It may be that generic
prompts are more beneficial than directed prompts (Davis, 2003)but it could also be that students
require scaffolding through directed prompts initially and that this scaffolding can fade to more
generic prompts once they master the appropriate self-directed learning skills (White and
Frederiksen, 1998).The approach used in this study was to use an open ended software
environment which did not itself provide structured prompting, complemented by written general
prompts supplied by the instructor, and face to face and written feedback. By examining what
effective self directed learners write about and the prompts with which they cue themselves, it is
possible to derive a set of prompts which may be useful to less effective self directed learners in the
future. These could be used in the design of self directed learning environments, software based or
otherwise.
METHODOLOGY The purpose of the study was to examine in depth the range of self directed learning skills in first
and second year students and, by contrasting the skills demonstrated by the most able students
with less sophisticated skills, provide some insight to other educators about how to facilitate self
directed learning. What research methodology would be most appropriate to this goal?
The present study uses qualitative analysis of written learning logs as a window onto students' self
directed learning skills. This is in contrast to some of the previous studies on supporting meta-
cognition in scientific inquiry learning which used a combination of qualitative and quantitative
approaches to relate learning gains to meta-cognitive strategy. As the emotional aspect of self-
directed learning and reflective writing in particular is considered important (Boekaerts, 2003,
Boud, 2001), it was necessary to find a methodology which would capture the students’ affective
experiences.. The methodology used in the current study is more closely related to that of
Luehmann, who used in depth qualitative analysis of blog posts to develop an understanding of the
16
potential benefits of reflective writing for teacher professional development (Luehmann, 2008). In
contrast to Luehmann, this paper studies the writing of a range of bloggers in an attempt to contrast
strategies used by capable and less capable self directed learners.
In a discussion of the methodological approaches used to study the challenging area of meta-
cognition, Anderson et al. argue that using a relativist-contextualist paradigm enables the
researcher to account for the complexity of the learning environment in which the learner is
situated, the motivation of the learner, and the nature of the subject matter(Anderson et al., 2009)
In their view, collecting data in an ongoing naturalistic manner can assist in developing a deeper
understanding of the interplay between meta-cognition and learning. They propose various
qualitative methods for data collection, although their work is focussed on meta-cognitive
behaviour in group discussion settings rather than the individual design task which is the focus of
this study. One recommendation of Anderson’s is to encourage students to select the episodes or
critical incidents which they would like to reflect on, so that learners are given some voice in the
research process. In the learning logs used in the present study, learners may choose what to write
about, and when, ensuring that the analysis reflects the matters which the learners themselves find
important. Interestingly, Anderson et al comment that taking part in the interviews is in itself a
meta-cognitive experience for learners, and that the same research methods which are used to
gather data to study meta-cognition can be of benefit to learners. Reflective writing in the form of a
learning log could also be beneficial to meta-cognition. Therefore, the learning logs used in the
present study served three purposes: an extremely rich source of qualitative research data, a tool
for learners to develop their meta-cognitive capacities, and a useful diagnostic aid for the author as
an educator who wished to assess her students’ directed learning skills.
(Brown, 1987)discusses in some detail how meta-cognition can be studied. She identifies problems
with using verbal responses and justifications as a measure of meta-cognition, namely that the act
of verbalising a task while doing it can be distracting and in some cases have a negative impact on
performance. She advises researchers to consider the temporal relationship between the report
and cognition in question. In the present study, the students regularly reflect on learning sessions
after they have taken place, thus reducing the likelihood that their performance will be disrupted.
She also makes the point that some forms of meta-cognition are not stateable; knowing how to do
an activity does not mean that you can explain how to do it to others. This is a limitation of
reflective writing as a methodology for studying self directed learning. Students may be capable of
directing their own learning, but may either not be capable of expressing it in a written form, or
may simply choose not to spend their time doing so. However, this trade-off is acceptable here,
given the benefits of discovering about the phenomenon in the learners’ own words.
ANALYSING THE STUDENTS' LEARNING LOGS
RESEARCH QUESTIONS The overarching aim of the analysis was to better understand first and second year computer
science students' capacity for self-directed learning as demonstrated through their reflective
learning logs. This was to some degree a benchmarking exercise to take stock of the range of skill
17
levels of the students, and the relative difficulty of different sub-skills of self-directed learning. The
analysis was also designed to uncover the writing strategies used by the more expert self directed
learners and the pitfalls encountered by the less expert students. This led to the following research
questions:
R1: Is there a relationship between the students' capacity for self directed learning as
manifested in the learning logs and their domain level technical abilities?
R2: To what extent did the students demonstrate self directed learning skills through their
blogs?
◦ a) Which subcategories of self-directed learning skills were found in the students'
blogs?
◦ b) Were some of these skills displayed more frequently than others? If so, which were
mastered and which require more development by instructors?
◦ c) How did the skills used change over the life-cycle of the project?
R3: For each self directed learning sub-skill: what are the hallmarks of a blog entry where a
student successfully demonstrates the skill, as compared to entries where a student has not
mastered the skill in question?
QUALITATIVE ANALYSIS METHODOLOGY Learning logs, with a total word count of 214295 words, were analysed in order to gain insight into
the student’s developing self directed learning skills, and other issues which were important to
them. The logs were written by the 113 students who submitted this part of the portfolio and gave
their permission for their files to be analysed. The logs were exported from the Blackboard virtual
learning environment in order to programmatically perform a word count. Microsoft Access was
used to store and search the corpus by category, and SPSS was used to perform inter-rater
reliability statistics.
Thematic analysis (Hayes, 2000)was used to broadly categorise the learning logs into the following
emergent areas of interest: affect and motivation; self directed learning; peer support and issues
relating to the Second Life environment. For simplicity, the author coded each excerpt with only one
category. Excerpts were generally between one and five sentences long but this varied according to
the meaning of the text; an attempt was made to code sentences which expanded on the same
thought together in one excerpt. There were 1935 excerpts coded. The focus of the current paper is
mostly on the evidence relating to self directed learning, but there is some overlap between the
help seeking aspect of self directed learning and peer support. Additionally, some learners who
reflected on their learning environment critiqued the Second Life environment, which is of interest
as this is an example of a self directed skill as identified by (Loyens et al., 2008). Examples of the
sorts of content which emerged under each heading are given below. These examples were given to
the second rater to train her on the coding scheme.
18
SELF DIRECTED LEARNING This category is used for blog excerpts where the student is thinking about their own learning. It
includes setting goals, planning work, evaluating their own performance, deciding on how to
respond to peer reviews, generating ideas, selecting between ideas, monitoring progress, managing
resources and in some cases critiquing features of the learning environment. Guidelines for
deciding whether an excerpt should be classed in this category are shown in Table 2. These
guidelines were derived by the author after an initial pass through the corpus, based on knowledge
about self directed learning from the literature.
In this excerpt the student… Estimates how hard or easy a task or subtask is Scales back task given task difficulty estimation Decides to switch strategy Decides on challenge for himself – involves assessing own abilities Sets a goal of what to work on e.g. what creature to build Plans what to do next time Reflects: looks back over a long period (e.g. semester) Decides where to get help from Improves on method to make product better Decides there isn’t enough time to do something Decides he need to spend more time on other modules/activities Decides which skills to develop Considers how to advance his own learning Decides on pros and cons of software to use Compares his own implementations/algorithms to those used in SL itself Identifies that the method he uses now might cause him problems later Decides how to respond to peer suggestions Identifies why something is hard Selects between creative ideas
Table 2. Coding guide for self directed learning category.
After the initial categorisation, each excerpt included in the self directed learning category was sub-
categorised as relating to planning( Table 3), evaluating( Table 5), monitoring(Table 4) or
help seeking. These sub-categories were chosen as a convenient way to group the main sorts of
skills which the students demonstrated as identified during the initial categorisation phase.
In the planning sub-category, the students plan their work including what to build and how to go
about it. This is usually at a lower level (day to day) than monitoring. In the help seeking
subcategory, the student recognises that she needs help and possibly decides what the best source
of help will be.
In this excerpt the student…
Generates own goals Decides what to build Compares/selects between ideas
19
Decide swhat to work on next time Plans subtasks in detail Manages a task Plans which method to use
Table 3. Coding guidelines for planning sub-category.
Monitoring is usually at a higher level than planning, when the student periodically reflects on their
progress so far.
In this excerpt the student…
Identifies that progress is slow Identifies that time is running out Decides to switch strategy for learning Realises that he doesn’t understand a concept Realises a particular way of learning is or is not effective Manages time or resources (e.g. how much effort to put into this module compared to others)
Table 4. Coding guidelines for monitoring sub-category.
In this excerpt the student comments on…
The difficulty of a task Their own performance A product they have created Whether an activity is challenging for them Whether and how to respond to peer review The learning environment (e.g. the software or text book) Whether the method they have used could be improved, or whether a method is appropriate
Table 5. Guidelines for coding evaluation sub-category.
AFFECT AND MOTIVATION Affect and motivation refers to the emotion or attitude which accompanies the student’s learning,
as well as their own reasons for wanting to complete a task. These can be either positive or
negative. The guidelines used to decide whether an excerpt belongs to this category are shown in
Table 6.
In this excerpt the student…
Mentions something is fun or enjoyable
Has a playful attitude to their work (e.g. mentions their pet in a humorous way)
Is proud of their work (this can be implied in the choice of words as well as stated directly)
Is disappointed in their work or progress
Feels confident or has high self efficacy
Lacks confidence or has low self efficacy
20
Feels frustrated, stressed, anxious, irritated or angry
Doesn’t want to complete a task
Is looking forward to completing a task
Has an affective attitude towards a particular learning task (e.g. it was particularly useful, or it was a waste of time)
Table 6. Guidelines for affect and motivation coding category.
PEER SUPPORT Excerpts were coded as peer support when the author interacted with peers using blog comments,
when they mentioned working with a peer in the labs, or receiving help from others. Peer review is
a special case, because if the student is commenting on the peer review and considering how they
will improve their work, this counts as self directed learning. The coding guidelines for deciding
whether an excerpt belongs in this category are shown in Table 7.
In this excerpt…
Student offers help to others as part of their own blog post
Student offers encouragement as a comment on another student’s blog post
Student asks a question using blog comments
Blog author replies to question asked by commenter
Student asks question in post
Blog commenter answers question
Blog author asks for opinion
Commenter give opinion
Student offers help/ resources in blog comment
Student offers advice on strategy in blog comment
Student describes learning with another student in the lab
Student describes getting or giving help to a peer
Student reflects on peer learning in general
Student reflects on peer review process, or thanks peers for reviews
Table 7. Coding guide for peer support category.
SECOND LIFE Finally excerpts in which the writer mentions their opinion or attitude to Second Life should be
coded in the Second Life category.
INTER-RATER RELIABILITY
21
Ten percent of the excerpts coded were randomly chosen to be categorised by a second rater, who
was first trained on a coding scheme written by the author and is summarized in the tables above.
This resulted in a Kappa value of 0.768 (p < .001), which is considered to be excellent agreement
(Fleiss, 1981).The contingency table (see
Rater 1
Motivation and affect Self directed learning Peer support Second Life
Rater 2 Motivation and affect 26 11 1 1
Self directed learning 11 75 2 0
Peer support 1 1 44 0
Second Life 0 0 0 10
Table 8) shows that the most common disagreement was between the motivation and self-directed
learning categories, which is not surprising, given (Boekaerts, 2003)view that affect is woven
throughout learning experiences.
Rater 1
Motivation and affect Self directed learning Peer support Second Life
Rater 2 Motivation and affect 26 11 1 1
Self directed learning 11 75 2 0
Peer support 1 1 44 0
Second Life 0 0 0 10
Table 8. Contingency table for inter-rater reliability for main categories.
The second-rater then sub-categorised a random ten percent of the excerpts coded by the first rater
as self directed learning according to instructions written by the first rater summarized in the
above tables. The kappa value was .551 (p < .001). As this is considered to be only “fair” agreement
(Fleiss 1981) the contingency table was examined (see Table 9). It was decided to collapse
evaluation and monitoring into a single category as the high disagreement between these
categories was reducing the reliability.
Rater 1
Planning Monitoring Evaluating Help Seeking
Rater 2 Planning 21 2 2 0
Monitoring 5 5 0 0
Evaluating 9 5 37 0
Help Seeking 0 0 2 2
Table 9. Contingency table for inter-rater reliability for self-directed learning sub-categories.
RESULTS The following sections present the results of the analysis, beginning with a breakdown of the marks
for each portfolio section which gives some indication of the relationship between student
performance on domain specific technical skills and more general self directed learning skills
(Research Question R1). Descriptive statistics for word counts in the student blogs are then used to
22
give an indication of how much the students wrote, and the relative frequencies of the coding
categories in the corpus is used to give a broad overview of the extent to which students
successfully carried out the reflective writing task (Research Question R2). Lastly, an in depth
description of the skills used by the students of different ability levels in the self directed learning
sub categories is presented (Research Question R3).
R1: RELATIONSHIP BETWEEN SELF DIRECTED LEARNING AND DOMAIN LEVEL SKILLS The average marks for each part of the portfolio course work are shown in Error! Reference
source not found. The average mark for the portfolio was 52%, indicating that the learners were
reasonably effective at reflective writing, but there was a range of abilities/effort (standard
deviation = 18).The averages for each of the portfolio sections are close to the overall portfolio
average (56%), and there are moderate correlations between marks in the learning logs and
scripting elements (r = 0.68), marks in the peer review and learning logs (r= 0.6) and a high
correlation between marks in the scripting and 3D modelling elements (r = 0.84). This suggest that
the students who are good at programming are somewhat likely to be good at the “soft skills” of
peer review and reflective writing, as well as the artistic skill of 3D modelling. It is not possible to
tell from this correlation whether the act of writing a learning log improves the technical product,
or whether those who are good at programming are also good at reflecting on it. Qualitative
analysis of the learning logs does indicate that some of the students who are most experienced at
programming are also capable of reflecting on their progress in a sophisticated fashion. However,
there are also some learners who are less experienced programmers who are capable of mature
reflection on their learning.
Peer review 3D modelling Scripting Learning log Average mark 54% 60% 56% 52% Standard deviation
23 21 21 18
Table 10. Breakdown of portfolio marks.
R2: THE EXTENT OF SELF DIRECTED LEARNING SKILLS AS DEMONSTRATED IN BLOGS
DESCRIPTIVE STATISTICS One hundred and thirteen of the students who submitted their learning logs as part of their
portfolio gave permission for them to be analysed (only three students withheld permission). The
learning log corpus totalled 214295 words. The descriptive statistics for total word counts, total
number of posts and mean word count are shown in Table 11. The kurtosis for total word count
was positive (3.64) indicating that more values were in the tails of the distribution rather than
clustered around the mean. There was positive skewness (1.8), indicating that there were a greater
number of small total word counts than large word counts. One student wrote considerably more
than his peers at 9382 words but this outlier was retained for analysis because the content was on
task and provided a rich source of data.
Mean Standard deviation Total word count 1953.39 1787.73
23
Total number of posts 10.96 7.42 Mean word count per post 171.85 129.56
Table 11. Descriptive statistics for blog word counts.
The word counts indicate that the students put a reasonable amount of effort into the learning logs,
producing on average the equivalent of a 2000 word essay spread out across several posts. On
average, they posted entries in their blog just under once a week. There was a wide variation in the
effort students put in to the learning log as measured by the word count; qualitative analysis
indicates that longer, more frequent posts are often associated with higher quality reflection.
RELATIVE FREQUENCIES OF SELF DIRECTED LEARNING SKILLS IN BLOGS Table 12 shows the percentage of students whose blog contained excerpts which were coded with
the broad categories, as well as the percentage of all excerpts which were coded using those
categories. The former indicates the proportion of the class whose writing displayed a category of
interest while the latter shows how prevalent the categories are across all the learning logs. Table
13 shows the same for the self directed learning sub-categories. It can be seen that the majority of
the students’ writing indicated some self directed learning was taking place (96%) even if this was
at the most basic level of evaluating their progress. This is not unexpected, given the assignment
instructions and the lecture material devoted to examples of how to write a good learning log.
Similarly, a high proportion of the student writing commented on affect or motivation (77%). Half
of the excerpts analysed were related to self directed learning, while around a quarter related to
supportive peer behaviour and just under a fifth related to affect or motivation. Around half of the
students commented on the Second Life environment, but this type of excerpt is not particularly
common in the corpus (5% of entries), mostly because the students commented on Second Life
when they first started to use it, or at the end of the module and not much in between. Only three of
the students’ learning logs contained no excerpts relating to any of the categories under analysis,
which implies that most students understood and carried out the task correctly.
Category
Number students
Count of excerpts
% of students
% of all excerpts
Motivation 87 351 77 18 Self directed learning 108 972 96 50 Peer support
94 473 83 24 Second Life Environment
54 104 48 5
Table 12. Frequency of categories
24
Number
students
Count of
excerpts
% of
students
% of all
excerpts
Planning 83 280 73 14
Evaluating
and
monitoring 103 642 91 33
Help
seeking 15 24 13 1
Table 13. Frequency of sub-categories.
Around half of the excerpts related to self directed learning, and 33% were related to the sub-skill
of evaluating and monitoring learning. 25% of the excerpts were related to peer support, reflecting
the module’s emphasis on collaborative activities such as peer review, pair programming, and
commenting on group blogs. A possible explanation for the low proportion of help seeking excerpts
(1%) is that these were categorised instead under the peer support category. However, it does not
explain why there are so few excerpts relating to help seeking from staff.
This simple frequency counting is informative only at a broad level to illustrate that the learning
logs gave evidence that the students were using the required skills. The following section gives
further insight into the differences between the more and less able self directed learners through in
depth qualitative analysis. Examples of coded excerpts are given to illustrate the students’ thinking.
The aim was to show the range of sophistication in self directed learning skills, rather than the
prevalence of these skills within sub-categories. Judgments in deciding whether self directed
strategies were effective or less effective were made by the author based on knowledge of the
literature but more importantly the context of the blog in question. The chronological accounts in
the blogs made a clear trajectory of progress, making it possible to link strategies with outcomes.
This approach is suitable in an exploratory study, but should be made more explicit and rigorous
for future quantitative studies.
R3: HALLMARKS OF SUCCESSFUL AND LESS SUCCESSFUL SELF DIRECTED LEARNING
SELF DIRECTED LEARNING – PLANNING
Self generated goals
The more able self directed learners in the class took charge of their own learning in the initial
stages of the project even before beginning on their pet projects. In contrast, the less able self
directed learners confined themselves to carrying out the instructions on the lab sheets without
creating additional goals for themselves, or extending the instructor specified goals to make them
more challenging.
For some learners, the Second Life environment was an invitation to explore the possibilities of
what they could build. They set themselves goals based on the affordances offered by the
25
environment. For example: “As for now I'm just getting used to all the basic functions and actions
on second life and making some very strange things (I've not decided what they are yet) that will
hopefully turn into something cool”. Observing strange effects within the environment caused some
learners to hypothesise about their cause, and experiment with parameters to understand the
underlying mechanism e.g. “Phantom [a Second Life setting]: Objects will pass right through other
objects; it may have more effects and didn't really look much. Interestingly can be combined with
physical [A second Life setting], but not sure what effect this produces yet. I will need to test this.”
Others set themselves mini-projects, such as building particular objects, in order to learn skills they
thought would be useful in their main pet project. In some cases these mini projects then evolved
into pets as the learners invested more and more effort in them. Even in these self directed projects,
some learners set themselves high standards: “I need to make it as robust as possible or it will just
irritate me.”
Some of the second year students appeared to take responsibility for learning the programming
language features from the beginning, rather than waiting for the class on particular topics. “It uses
states. I did this as a test to try and familiarize myself with them.” One student explicitly wrote
about this “Again a bit of a short log as I didn't really do that much I felt challenged me this session,
I think I'll try to take learning LSL into my own hands in future so that I can advance at my own
pace.” The more experienced students were also capable of extending and modifying lab tasks to
the correct level of challenge for their abilities; “It was rather simple to do (based loosely on last
week’s lab), so I expanded it and made it more interesting”.
Generating and selecting ideas Deciding on which pet to create required the students to generate possible ideas and then select
between them. This is a necessary skill for creative projects and requires the ability to evaluate the
potential difficulty of a problem as well as the potential value of the solution. Hall-marks of less able
self directed learners were not evaluating the feasibility of the idea before investing time in it, and
not successfully matching the difficulty of the project to their current abilities.
However, many of the students made some effort to set themselves a project of the correct
challenge level. One first year student commented “To choose what pet to design and build for this
course in Second Life, is like choosing a car: it has to work when you need it to; it has to be easy to
use (or in this case, to design).Ok, so not really easy, but more of a case of being within my ability.”
Another student wrote about her choice of idea “this seemed a good challenge for me to do and
although seemed quite hard I knew it was achievable”. Naturally, the effectiveness of challenge
setting depends on the realism of their evaluation, as well as their confidence. For example,
although Student M wrote: “I'm trying to make my cat follow me, and if I could, I would make it run
from other people, but that's being a little ambitious for me”, he was incorrect in his estimate of
both the difficulty of the task and his own abilities. In fact, throughout the learning log, this student
consistently demonstrates lack of confidence and pessimism in spite of performing well.
There was recognition from many students that their decision about what pet to make was related
to their familiarity with and development of skills in Second Life, and that as their skills grew, their
ideas would evolve. “I chose my pet and it is going to be a starfish. Since I am not a talented
26
draughtsman the animal is rather simple to draw. However I think it has got a big potential and it is
possible to develop it into a much more complicated project if necessary. I just have never played
with 3D graphic before and on the one side I don’t want to overestimate my abilities; on the other I
would like to leave a possibility to create something more complicated if I found 3D designing
interesting.” Some students deferred working on some aspects of their pet until they developed the
skills they needed ““I plan to return and try these examples when I am more comfortable scripting.”
or deferred making a final decision about their pet until they had more experience: “I hope to be
more sure once I get a hang of the programme and figure out what my capabilities will be.”
The final decision about which pet to choose was also influenced by personal preference such as
liking certain animals, wanting to explore a certain technical topic and being attracted by the
complexity of the topic. Some learners were influenced by external factors such as whether other
students were working on the same type of pet, or whether they thought it would demonstrate the
sort of features required to get good marks. Some students examined pet possibilities in some detail
before selecting one. “I decided to do the school of fish over the flock of birds due to the simplicity
factor (the flock of birds would be way too hard to create in terms of design (the feathers and body
shape would be extremely difficult) and animation (the fish would be situated in a smaller confined
underwater area where as the birds have way more room and therefore added complexity).” This
demonstrates the level of analysis of task difficulty of which the more successful students were
capable.
Having decided on a pet idea, the more able self directed learners displayed flexibility in altering
their ideas according to their progress, for example “Unfortunately my idea didn't work. My
knowledge of SL is limited at the moment to get my idea to work as I wanted it to. So I decided to
adapt my original idea”. Learners who did not mind scaling back their ideas early tended to get less
frustrated. In comparison, a technically gifted student put in a huge amount of effort into making his
dinosaur dance and was disappointed when he eventually decided that it would not be possible to
achieve given the technical limitations of Second Life. This situation presented an interesting
dilemma to the instructor: at what point is it sensible to intervene to prevent ambitious students
pursuing challenging goals to protect them from frustrating failures? In this particular case, the
technical limitations were not initially apparent to the instructor because the student’s idea was so
novel. It may have worked; but it was high risk.
Planning learning The students planned their work at two levels: at a day to day level of deciding what to work on
next; and at the higher level of considering the strategy they should use. Unsurprisingly, the latter
skill was considerably less common, and was demonstrated by only the more able self directed
learners. Examples of strategies include explicitly deciding to learn more about Second Life before
deciding on a pet, deciding to use trial and error, and favouring online tutorials as well as
experimentation. One student wrote about his strategy of choosing a difficult problem: “I have
made progress on the animal idea front, I want to do something interesting, interesting usually
means difficult. I want to push myself on this course, take chances and risks, I wouldn’t be happy if I
played it safe and got a C. I want an A!!! Truth is to succeed risks are part of life.” Another student
thought in some detail about the development life cycle of the pet as follows: “I started thinking
27
about the development of how I am going to create the pet. I think firstly I need to learn
programming during labs rather than during the week, this may sound obvious but it does take
time away from modelling the pet, and scripting it. The creation of the pet itself, will have to going
through a few development phases. I was thinking firstly what the pet should look like which is
simple because I have a photograph of it already. Next, I should think about what prims [3D
building blocks] will be needed to create the pet, again this maybe more complicated than it sounds
since we need to make it all work all at the same once it is scripted. Lastly, what textures will be
used and how much memory it may take up. These are just basic considerations and I'm likely to
encounter things that I had not considered.”
At the level of day to day planning, the students who made specific plans to accomplish small,
achievable goals tended to make more progress, particularly if they planned a time to complete
their tasks. For example “Going to work my way through from chapter three to six before the next
lab hopefully. Also, going to draw some ideas down onto paper about my pet.” Students sometimes
finished their blog entries with “to- do list” style notes on tasks they wished to complete, or topics
they needed to learn about, demonstrating that they were able to decompose the tasks into
manageable parts. Student who made vague plans (as in “Finally got the second life book and plan
to work on it now that I can look up on something that will help me go along.”) were less likely to be
able to stick to them. For some performance oriented students, it was important to direct effort
where they thought marks were available. “Next time I would probably hope to continue working
on the scripting as I think it will be where I'll achieve the majority of my marks for this module.”
This is discussed further in the task management section.
An interesting facet of the most able learners’ ability to plan their learning was their attention to
selecting between methods of achieving a task. This occurred in students who were more
experienced in project work, either because they were in second year or because they had
previously volunteered on open source programming projects. As an example, consider the
deliberations of Student J: “For this reason I will soon be looking into other methods of modelling,
such as modelling in Blender or 3ds Max, and then looking for methods of which to import these
models into the Second Life engine… am still very reluctant to go down the 3DS Max route as this
will probably eat a lot of time on its own, and may leave less time for the scripting and my plan of
what to do after the scripting is completed… If it does works I will search for other free software
plugins first, then look into how to get the program, maybe I can get some sort of discount through
the University. Will see if the plan works first and there are no other options before launching into
this though.” Student J has researched multiple possible solutions to a problem he had identified,
realises there are time implications which will impact on the rest of his project and comes up with a
back up plan. At the other end of the spectrum, some students had trouble finding even one possible
way of solving their problems or got stuck if their first method failed.
Task management At a basic level, the students had to decompose the task of making the interactive pet into
manageable sub tasks, and complete these before the final deadline. As is common in software
projects, many students did not complete the task in time and submitted pets with functionality
which fell short of their original plans. Although this was sometimes caused by lack of experience in
28
generating an achievable goal, lack of task management skills were often to blame. Students who
carefully analysed the task into sub tasks and set themselves interim deadlines were more likely to
succeed.
A second year student, who seemed determinedly in control of his learning from the beginning was
able to advance plan the stages of his project with some rigour: “The actual physical design of the
fish will start at the end of this week (after the final code has been tested and has been logged
correctly).” Unlike some of the other students, the behaviour of the pet has a higher priority for him
than the appearance; he realised that the two could be separated until the final stages of the project.
Student J again demonstrated his experience in managing his own learning by planning not only
interim deadlines for task completion, but also specifying the quality of the product at that stage.
“This means that I will have tomorrow and Friday to work on the head, so hopefully I can have a
complete first draft model of the dinosaur by Friday, so I'll post all the screenshots when I get the
first draft completed. The head will probably be difficult initially to model, so I might attempt a 50%
likeness attempt for first draft in a way that I can keep the same shape but give more additions on
my next model shot to increase realism.” It is not clear, though, how one would measure 50%
likeness to a dinosaur head.
Another experienced programmer decided that he needed not just to solve an isolated problem, but
that a more systematic framework was required for his code: “I need to design an architecture for
making this work smoothly.” This example is interesting because it indicates that the student was
not merely composing the task into subtasks, but was thinking about it at a more abstract level to
produce a generalisable solution.
As an example of a student recognising the dependencies of the project, relatively rare among the
blogs, is from Student Cr. He is writing about the best time for documenting his source code:
“I intend to do [it] as close to the dead line as I possibly can …. … the code that I have now is
probably not going to be the final version so if I comment it now then I will probably end up
changing it later. ….”
SELF DIRECTED LEARNING – EVALUATING AND MONITORING
EVALUATING OWN PERFORMANCE OR PRODUCT One of the most common evaluative practices was when the students evaluated their own progress,
or the work they had completed in a session. This varied from a simplistic statement of whether the
session went well or poorly (“Today I learnt some new stuff”), to more sophisticated analyses of
strengths and weaknesses, such as “My strengths were, that I like everything to be perfect, so when
building my snowman I ensured that objects were aligned correctly and that each object was in the
correct place... My weaknesses was: that at first I couldn’t (when copying a prim) then resize it but
keep it centrally aligned with the object beneath it. After looking through the book and checking
everything over and over I managed to realise that I just had to check a box in the edit table. So my
problem at the moment is that I don't fully understand all the actions within the edit table. So for
the future I need to learn all these actions to save me from confusion and to allow my creation to be
29
the best that it can be”. Some students evaluated the effort they put into a task (or in the module),
describing their “hard work” or identifying that they were “lazy”. Taking this a step further, one
student thought of several possible reasons for his perceived lack of progress “I didn't manage to
make them change colour.. maybe I didn't try enough, maybe there where a lot of distractions and
noises, or most probably I'm a really slow learner.!” Later in the module, this student was relieved
that he became more successful and noted this was when he was feeling less stressed.
Demonstrating a certain amount of meta-cognitive knowledge, one of the more mature learners in
first year remarked on how his habitual strategy was not working “For the other two tasks, I just
fiddled with the code I produced for the previous tasks. ..I tend to take what was called at school
"The muddling through approach". This takes a long time, and it’s a bad habit I need to break out of,
but it means that if I don't get it right, I at least know that certain ways are wrong.”
Here student F has identified that although she is weak in a particular area, the practice has
improved her understanding, thus demonstrating she had a meta-cognitive experience (Flavell,
1971). “My scripting skills aren't so good and so this has made my understanding of IF statements
much better”. Another student’s comment illustrated that his meta-cognitive knowledge is lacking
“Despite 3 of us working together we where struggling to figure out how to solve the problem
despite knowing exactly what we where trying to do (Yes, very contradictory).” This student is
apparently unaware that being able to specify a problem is distinct from finding a way to solve it.
Even among the less experienced first year learners, there were analyses of which aspects of the lab
exercises they had successfully completed, which they had found hard, and which required more
work, e.g. “Today’s lab was challenging, a bit difficult but not impossible, we ran out of time before
we could finish all the tasks. I think I just need more practice with the scripting.” A few weeks into
the semester, the more aware students often commented on their progress since beginning e.g.
“The scripting isn't as bad as I expected, I'm not finding it too difficult to think of what I want to
enter now. … I'm progressing slowly but surely”. The students were often critical of what they
achieved in the labs. However, if the less experienced learners did solve one of the exercises it
appeared to increase their confidence and motivation, as one would predict. The more able
students were often not happy with the limitations of their solution, even if they achieved what was
asked in the lab exercise. It was quite common for the students to identify extensions to the
exercises, but it was less common for students to actually implement them. The experienced
programmers, particularly in the second year were occasionally slightly contemptuous of the easier
exercises (“The code is also rather easy to implement and is nothing to strenuous to the
programmer”) but not all students in this group then took matters into their own hands to set more
complex goals for themselves.
We explicitly asked the learners to evaluate their own progress in order to decide which lab
exercises to attempt in a lab session in week 5. We offered either a set of practice exercises
designed to consolidate existing programming skills, or some more advanced exercises teaching a
new set of skills. Typically the extra help examples were chosen by those most in need of them and
were regarded as helpful. “I found the first two examples quite easy which gave some confidence to
go on and try the loops. I think having extra practice examples with extra lab class on a Tuesday will
help me greatly and I hope to be able to transfer those skills over to improve my wasp’s
30
interactions.” Our experience from this suggests that offering optional practice exercises for less
confident or less experienced students is worthwhile and we intend to develop further materials of
this sort for next year. The advanced exercises were useful to the more experienced learners at this
stage, and helped them to make progress with their pets.
When brain storming ideas for pets, some students evaluated the quality of their ideas, in
comparison to the ideas they heard about from other students, how novel the ideas seemed, or
whether the prospective pet would exhibit enough interesting behaviour to get high marks. As the
pets progressed, the learners would evaluate them regularly, comparing what they had built to
what they imagined. This varied from a simple statement about whether it looked good or not, but
more often was more detailed, as in “I spent a lot of time going through all the different textures to
try and get something that suited a bit more. …. Also, for the skin of the turtle, I have used leather
fabric texture. I think this is quite good as a turtles skin is quite leathery to touch. I really need to
just get the colouring right to show the patterns off to their full. I feel that it is still just a bit dull at
the moment, so I want to try and brighten it up a bit, bring it to life a bit more. ...” Students were
sometimes not happy with the appearance of their pet but accepted this as a result of their
perceived lack of artistic ability. The visual nature of the pet usually made it easy for them to judge
whether it was good or not. However a student with low confidence was unnecessarily hard on his
work, writing “Must make [my pets] do something interesting by Thurs. They're really disgraceful
as they are now.” In fact, his pets were nominated for the Virtual Crufts ceremony and were well
regarded by his peers. He was unable to accept these positive views of his work, however.
After the peer review session, the students were explicitly asked to reflect on whether they would
take the advice offered by their peers. This required them to evaluate the quality of their peers’
advice and decide whether it was feasible to act on it in the time available. If they decided to take
the advice, they had to decide how best to put it into practice, although the better peer reviewers
also offered suggestions of how the advice could be implemented. This exercise required some self
regulation because they had to compare their own assessment of their work to their peers’ and
decide whether they were consistent. Much of the time, the students were aware of the limitations
of their work and agreed when this was pointed out by their peers. However, it was common for
them to state there may be difficulties in implementing the suggestion due to time constraints, or
technical reasons which had not occurred to the reviewer. For example “I think Student L has some
fair points… Movement of the legs will be really hard, but I could try it. It's not the animation of the
prim, it's the fact that there's actually two prims on the leg, so I would have to get them moving in
unison.” Less confident students sometimes declined to follow suggestions because they believed
they lacked the skill to try citing reasons such as “llMessageLinked seems a little tough for me.”
Some students rejected suggestions of their reviewers because they did not fit in with their original
vision for the pet e.g. they wanted a particular colour scheme for aesthetic reasons.
We suggested to the students that the last post in their blogs should be a reflection looking back
over the module as a whole. This was useful to gain insight into the students’ learning trajectories
from their own point of view. The students generally admitted to having learned something –
unsurprisingly as they knew they were being assessed on the log – even the experienced
programmers among the second years. For example, one student wrote “The things I reckon I have
31
learned or improved on most of all though aren’t so much technical as social as I think this course
really helps boost that particular skill set by conferring with fellow students about each others’ pets
and what the plans for them are and also the fact that the labs are quite difficult it would be foolish
not to socialise and get different perspectives on how to do things”.
The first years tended to have more straightforward reflections on the content of the module as it
was new to them. In the following excerpt, the student acknowledges that he could have done
better and notes an area he found particularly hard. “I certainly feel that this work is no where near
my potential but I hope I pass so I can have a fresh start at the beginning of 2009 and get stuck in.
Overall this has been quite a fun module although the event driven coding was a bit hard to get my
head round especially getting thrown into it at the beginning. Another cool module next semester
please! ” Other first years seemed to have a vague feeling that they could have worked harder but
did not analyse this in much detail e.g. “Regrets of course are that I should have started earlier,
worked harder etc”. Even some students who were pleased with the outcome of the project had
ideas for improvements they could have made if they had more time “Despite the fact my penguin
3d model is not perfect, I do like it and I think that it actually looks like a penguin…To sum it up, I do
think that I could have done better. I see lots of ways to improve what I've done.” One learner, who
chose her pet based on her perception of her abilities commented how she might do it differently in
the future “If I was to start my pet again, I would definitely try and add more complex script maybe
to get it to walk/follow you around, I was very interested in getting my penguin to dance but
unfortunately this was too complicated at the time for me to do.”
In the following excerpt, the student was pleased with his own product but compared it
unfavourably with those of his peers “No where near the standard of some the awesome pets I have
seen on the island but I knew I wouldn't be able to match them. Even so I'm quite chuffed at how he
turned out, looks as if he came straight out of Final Fantasy (almost!) and that was the aim.” The
more able self directed learners had goals relating not just to the end result of their pet, but also to
the standard of engineering with which it was built. For some students, the elegance of the code
was important, as student H reflects here: “but I managed to greatly shrink down the amount of
redundant code and make it easier to understand what the hell is going on so now it is pretty and
well written. A good day.” The appropriateness of the algorithm, code efficiency, readability
succinctness and robustness was also important for students with more experience. This also
applied to artistic students who set themselves high standards for the 3D modelling task: “Most
recently I redone my dragon’s feet, because these were the first thing that I made when I started the
dragon, these were not up to the same standard as the rest of it, as I was making the dragon I got
more experience and learnt how to do things faster and a lot neater so I decided to change the feet.”
EVALUATING/ PREDICTING TASK DIFFICULTY At the beginning of the semester when the students were inexperienced in Second Life, it was
common for them to wonder or attempt to predict how difficult a task might be. Some students with
little programming experience were apprehensive about the difficulty of the module in general.
Some students made a habit of comparing the actual difficulty of the task they completed to how
they expected it to be. For example, Student R wrote: “Reading through the book as I went along
32
helped a lot, the snowman was not such a problem to build in the end….I had a quick attempt at our
previous lab last week and was a little concerned, thinking it was much harder than it actually is.”
Similarly, Student A commented on how his estimation of difficulty had been distorted by stress:
“today I wasn't stressed and I actually did it by my self. It’s amazing how easy it really is, I guess just
the thought of not knowing restricts you, you don't have to know everything tho to do something.
You can do things even if you don't know all the codes by heart, sometimes you just need to use
your logic”.
Occasionally the students underestimated the task difficult due to their inexperience, which
resulted in some disappointments. One student’s uncertainty about the feasibility of the pet is
apparent in the following excerpt: “Unfortunately I think I’m grasping for something out of reach
but never lose hope right?”
One student had the insight to realise that his own effort, rather than his pre-existing knowledge,
would impact on his marks. After previous entries where he worried about his lack of programming
skills, he wrote “I thought that at first this would be where I would do not so well in this module
however after this lecture it doesn’t seem as bad as I had initially thought. As long as I put in the
effort it should be ok. I think this will be the deciding factor in how well I do in the SL module.” This
is in contrast to many of the less experienced programmers who had lower self efficacy.
MONITORING A hallmark of the more effective self directed learners was their ability to monitor their own
progress, and to replan their work based on that. Some students who thought they were weak on
programming decided to direct their time to learning that aspect of the module, thereby suspending
their work on the easier task of 3D modelling. Here a student notes that he might need to change his
original idea if he is unable to solve a difficult problem after spending some time on it “I've been
trying to figure out how to make the pet follow whoever touches it, but I just can't get it. Another
few days trying to find out how to do it, and if I still can't get it, I think I'll have to call it quits and
think of something else to make the pet do.” Another student reluctantly had to give up a favoured
idea after exploration within a set time scale did not yield results: “this meant I either had to
redesign the whole chocobo [his pet] or abandon the idea of riding it, it was lose-lose either way but
rather than risk taking the time to recreate the chocobo only for the ride method not to work again I
had to abandon the idea.”
The impending deadline in week 12 was the trigger for some students to re-evaluate the strategy
they were using, and set themselves interim targets to better enable them to monitor progress. “I
am really starting to get concerned as I do not see myself finishing the pet on time. I've decided to
ration my time carefully and set targets as to what should be finished when. Taking on the advice
from my peer review I decided to create a limit as to how much Doggie can eat and when it can go
to the toilet. I set myself a target of two hours for this.” Another sophisticated self directed learner
deliberately decided not to estimate a time for task completion because “My experience with
modelling has taught me that it can be a lot harder than it looks- so I'm not setting any kind of
estimated date when I think this import script will be completed and working.”
33
Some students, having noticed that their progress was not as they hoped (“Ughh. Wish I hadn't
completely ignored this project over the past few weeks in favour of other coursework.”), decided
to put more effort in to the module (“I feel that I need to work quite a bit more in my spare time,
finding tutorials, experimenting and so on to be able to achieve the level of complexity I want my
pet to achieve.”) In contrast, another student who realised he was short of time seemed to take a
rather detached attitude to the quality of his final product: “Due to the now apparent time
constraints I have not so wisely inflicted upon myself I’m not in much of a position to be fussy on
what I produce for the final result. I am curious as anyone to see how it all turns out. Will try to
keep an accurate log during my week long adventure to produce my cow/parrot/pet thing.”
Another student in the same position panicked and decided to work late at night to get more work
done. He realised later that this was not perhaps the wisest strategy for him because the standard
of his work dropped: “Last couple nights I have been trying to perfect my Quiz script. Last night was
a complete waste of time I feel all I did was mutilate what little I had into a mess that didn’t work in
the slightest - so I scrapped it today.”
HELP SEEKING The students mostly wrote about seeking help form their peers, and occasionally from the lab
helpers or staff. There is considerable evidence from the learning logs that the students formed an
informal supportive peer community, providing technical help in the labs, answering questions
asked by the authors of other learning logs, posting advice and tips on their own blogs and
providing support and encouragement through the blog comments. This was encouraged by the
staff, and more formal opportunities for peer learning were also designed into the module in the
form of paired programming during lab classes and an assessed peer review exercise. The lecturer
also requested and arranged for more experienced students to provide peer support during lab
classes in addition to staff support.
The reasons for seeking help included not being able to complete a task, identifying that a script
was poorly constructed or realising that progress was very slow.
One of the students with weaker scripting skills wrote a post noting that he needed help fairly
urgently, after falling behind. “To be honest, I’ve missed a couple of labs, but have tried to do the
work myself, but to no avail. At first this scripting business didn’t seem too hard, but how oh so
wrong I was. I am going to have to get proper help to get me up to speed! It doesn’t help that I had
no previous programming experience either, so I’ve not done myself any favours. It’s my own fault
and well, I’d be a fool to say it wasn’t. Hopefully from here on in things can change and I can actually
get somewhere with the project.” The student later mastered the scripting skills he needed, and
reflected “I can sort it all on my lonesome now as well so I’m not such a huge failure anymore.” The
fact that he refers to himself as having been a failure suggests that his confidence was considerably
dented by the time he did ask for help, and that he might have had a more pleasant experience if he
had asked earlier.
A rather misguided observation about not seeking help comes from a second year student who
wrote “I'm not sure whether to ask someone on Monday or tomorrow if they're online. I don't think
asking is a good idea because you can't do that in a professional job”. This student perhaps did not
34
appreciate that being an IT professional requires constant updating of skills, so learning form one’s
colleagues is part of the job. Another student commented that she would ask for help from her
classmates if they were available but when she figured it out for herself it gave her satisfaction. She
also noted that she would only ask help from staff when it was crucial because she didn’t want to
get too dependant on them.
EVALUATING THE LEARNING ENVIRONMENT Any creative task is constrained by the environment the designer must work within. There were
many comments from students relating to Second Life in this respect, and although this is not the
focus of the paper, cases where the student seriously evaluates the features of Second Life with
respect to their learning are of interest. Many of the comments were affective, relating to
frustration about technical problems or excitement about the initial novelty of it. However, the
students who had some background in graphics or other programming languages were often
critical. There were some criticisms about the quality of the 3D products which could be modelled
within it “After putting it together with the legs I resigned myself to the fact that it looked okay by
Second Life standards”. It is interesting that none of these students identified why the technical
problems in Second Life are inevitable given the ambitious nature of the system. The technical
reasons for this could be studied in a more advanced module.
However, it is fair to say that technical problems with Second Life did cause some difficulties. The
more experienced students were able to predict aspects of their designs which might run into
difficulty and plan ahead “There are going to be problems upcoming with packages in the island
being too full- if my birds are flying around the island and meet this problem they will just stop
dead still... so that's on my "to fix" list...”
One particular student was passionate in his dislike for the LSL language, citing a couple of features
he found irritating. However, it is worth noting that the features he mentions are related to
personal preference and does not necessarily represent a mature comparison of languages. “If the
exercise has taught me anything, it's that LSL is an absolute abomination of programming
languages! I know it's not meant to be OO but it could do with having a .length property for 'lists'
(read: arrays) and I mean what on earth is up with not being able to access items in a list by their
index using the square brackets!?”
Another student was able to predict the difficulty of the programming task in the module based on
his experience of other languages “Mainly looking over some scripting today and getting familiar
with it again. I don't think the actual programming itself will cause that many problems, the syntax
of the language is pretty straight forward and if you have a good reference with a list of all the all
functions then it should be fine”
In terms of a learning trajectory, student MZ had an interesting development. He started off with a
very critical attitude to Second Life, and found the programming too easy. Once he started working
on his pet, he realised that he had the freedom to set himself more challenging goals, and in doing
this he explored technical issues in Second Life in considerable depth. “Finally, scripting in SL was
really a great experience. I really regret I didn't start making more serious scripts earlier. I think the
environment has a big potential in spite of limits. Actually these limits reminded me about limits
35
when I used to make programs in Pascal and basically you had only 640KB, which you had to share
with OS and drivers”.
RESULTS SUMMARY AND ADVICE FOR EDUCATORS The exploratory findings from the qualitative analysis of the learning log corpus give a benchmark
of the level and range of self directed learning skills among this set of first and second year students
on a technical degree course. The findings suggest avenues by which educators can assist students
in improving their self directed learning, and help to identify possibly fruitful areas for further
research in the learning sciences.
Research question Summary of findings R1: Is there a relationship between the students' capacity for self directed learning as manifested in the learning logs and their domain level technical abilities?
Yes. There is a moderate correlation between scores for reflective writing and interactive pet components of the portfolio.
R2: To what extent did the students demonstrate self directed learning skills through their blogs?
96% of the student blogs demonstrated evidence of self-directed learning, although there was a wide range of abilities within this.
a) Which subcategories of self-directed learning skills were found in the students' blogs?
Planning, evaluating and monitoring, help seeking.
b) b)Were some of these skills displayed more frequently than others? If so, which were mastered and which require more development by instructors?
Evaluation was more common than planning. Students would benefit from assistance in planning more frequently and in evaluating progress on the project as a whole at a higher level. Help seeking (from instructors) was not common.
c) How did the skills used change over the life-cycle of the project?
There was some tendency for the frequency of blog entries to decrease in the middle of semester. Early entries reflected on progress and performance in individual lab sessions, whereas later entries focused on evaluating the product. Success self directed learners used blog entries mid-semester for planning and task management.
R3. For each self directed learning sub-skill: what are the hallmarks of a blog entry where a student successfully demonstrates the skill, as compared to entries where a student has not mastered the skill in question?
See Table 15
Table 14. Summary of answers to research questions.
There was a relationship between domain level skills required to program an interactive pet with
self-directed learning skills as demonstrated in the learning logs. This is consistent with findings by
36
previous researchers who have found that improvements in self directed learning skills are
associated with domain level improvements (Vrugt and Oort, 2008, White and Frederiksen,
1998)By and large, students who did well at making their pets also did well at writing reflective
logs about this process, although it is not possible to tell whether there is a cause and effect
relationship here. Further quantitative research is required to establish whether students who
regularly write learning logs outperform students of matched prior technical ability who do not.
There were some exceptions to this relationship between domain and self-directed learning skills,
for example in the cases of mature learners who had little prior programming experience or
experienced programmers who were impatient about having to write about the process as well as
actually carrying out the task. As (White and Frederiksen, 1998) note, students can find too much
self assessment burdensome. It may be that students who possess more skills in the domain have
also previously learned the requisite self directed learning skills to help them manage projects of a
similar nature and do not require extra scaffolding from the learning environment. The group most
likely to benefit from scaffolding in self directed learning skills include those who find the domain
difficult to master and therefore find it necessary to re-plan their work around failure; prompts
which might help such students develop the skills are suggested in Table 15. The group of students
who do not actively engage with the course material is difficult to help because they neither put in
the effort to work on their project nor evaluate their lack of progress in their learning logs. A
possible solution to this is for educators to be automatically notified when students have not posted
to their learning logs for a pre-specified period of time. This would enable them to be proactive
about trying to encourage the students in question to re-engage with the material.
Evidence from the learning logs indicates that most of the students were capable of reflective
writing which gave an account of their self directed learning strategies. The qualitative analysis
indicates that the students set their own goals, critically evaluated the learning environment,
identified gaps in their learning (Loyens et al., 2008) and engaged in the self regulatory behaviours
of planning (Brown, 1987), evaluating success (Boekaerts, 2003), predicting outcomes and
scheduling strategies (Brown, 1987). There is also some evidence relating to the students’ meta-
cognitive knowledge, meta-cognitive experience (Flavell, 1979), and some indication of the
students’ self efficacy (Coutinho and Neuman, 2008), persistence and willingness to put in effort
(Vrugt and Oort, 2008). The level of self directed learning ability demonstrated in the writing varied
widely, however. Sub-skills relating to evaluation were most commonly used by students, whereas
planning sub-skills were less frequently employed. Students were more inclined to evaluate their
task performance directly after lab sessions and less likely to evaluate their progress on the project
as a whole. There was also a general tendency for students to blog regularly early in semester (up
to around week 5), blog less frequently for some weeks, then blog again in response to peer reviews
at week 10, followed by a final reflective blog at the deadline in week 12. There are several possible
reasons for this. One is the general mid-semester “dip” in student attendance and motivation which
is commonly noted by staff in the department. It could be partly related to assignment deadlines for
other continuously assessed modules which require the students to direct their attention away
from the Interactive Systems module. It could also be that as the lab exercises grew more difficult
throughout semester, the students were less likely to reach the final question on the lab sheet
which was a prompt to blog about their learning during the session. That is, they were spending
cognitive effort on working on a task itself rather than meta-cognitive effort on reflecting on their
37
success at the task. In spite of these possible pressures not to blog during the middle of semester,
the more effective self directed learners did use their blogs as a way of monitoring their progress on
the interactive pet project. This enabled them to plan how to use their time effectively for this
module and take steps to put their project back on track if it was not progressing as intended. In
contrast, those learners who did not plan were more inclined to panic towards the end of the
semester when the deadline grew nearer and they realized that they had run out of time to
complete the pet that they originally intended to make. This suggests that the students could
benefit from more structured prompting from staff to plan and monitor their progress around the
middle of the project, and some help in managing their time given the deadlines on other modules.
The learning logs suggest that most of the students actively engaged with making their pets and
found the task intrinsically motivating. The learning logs were also a good source of data on student
attitudes and affect, which often coloured meta-cognitive activities such as performance evaluation,
as predicted by the work of (Boekaerts, 2003). The affect/ motivation data from this corpus merits
further study from the perspective of promoting student engagement, an issue which is particularly
important for first year students. We need to know more about which features of an open ended
creative project lead to positive and negative emotional experiences, and the most effective ways of
assisting students in managing these contradictory feelings.
A gratifyingly effective component of the pedagogical design was the emphasis on peer support,
particularly through blog comments. This collaborative, communicative affordance of this form of
social networking has been predicted to have educational advantages by previous researchers,
particularly by Minocha (Minocha, 2009b).The students provided helpful, encouraging and
constructive comments on each other’s writing far more frequently than the staff had time to, thus
suggesting a pragmatic way of providing affective support for first year students. This is
particularly important as there were not many examples of writing about help seeking behavior
otherwise. Either the students did not often choose to write about occasions when they got help
from sources such as teaching staff or they did not consider this as a strategy to help them when
they had difficulty. The instructors were certainly very busy during lab classes but it is possible that
there were too many students to support them all effectively during class time. Furthermore,
students working independently on their project work are likely to encounter problems out-with
class time and with such large class sizes, it is not possible for the teaching staff to systematically
provide individual help electronically.
The qualitative analysis in response to R3 gives an indication of the contrasting approaches of the
effective and less effective self directed learners identified in the blog (see Table 15 for a summary).
It also suggests ways in which educators could promote effective self directed learning skills, along
with some directed prompts which could be appropriate at different stages in a creative project.
The prompts could be incorporated into lab worksheets, used verbally by instructors during
classes, or (after further pilot testing) be designed into a more structured writing environment
similar to those developed by previous researchers (Loh et al., 1998, Scardamalia and Bereiter,
1991, White and Frederiksen, 1998). Whether more directed prompts would be more effective than
the existing more generic prompts is an open research question. Davis’ work suggests that in fact
the open ended nature of generic prompts can be more effective than educator-generated directed
38
prompts (Davis, 2003), in which case the flexibility afforded by blogging tools might in fact be more
appropriate to this learning situation than a more structured software tool. An environment in
which prompts start off directed and become more generic as students master the requisite skills
might be a sensible middle ground.
The educator has an important part to play in assisting students who are inexperienced self
directed learners in gaining the skills they need to manage an open ended creative project for
themselves. As (Scardamalia and Bereiter, 1991) argue, the educator is trying to increase the
student’s autonomy and so should not be too prescriptive in structuring the learning environment.
However, there are some broad issues which educators should be aware of in support the self
directed learning sub-skills of planning, evaluating and help seeking.
In the early stages of a creative project, the educator can encourage students to explore the
possibilities of learning environment in a playful way and take responsibility for setting their own
learning goals which are appropriately challenging for their current ability level. When helping
learners to select between different creative ideas, the educator can offer useful advice about the
relative difficulty of achieving these ideas given the current experience of the students to help them
to avoid disappointment and frustration later. It is difficult to achieve a balance; learners ultimately
learn how to estimate task difficulty for themselves, but initially they require some scaffolding from
someone more experienced as they have no way of knowing how hard a task is to achieve within a
new learning environment. The students can be advised to monitor the difficulty of the task for
themselves once they start it and counseled on how to moderate the task should it prove to be too
hard or easy.
Self-efficacy is a factor here: some students may require encouragement to try a more ambitious
project if the educator has reason to believe that the student’s technical abilities outstrip his
confidence. As the project progresses, the educator can prompt the students to plan their learning
through finishing each session by deciding on the next learning steps they should take next session.
At a higher level, the learners should also be encouraged to set themselves regular deadlines
(perhaps on a week by week basis), identify the dependencies between tasks and use a strategy of
iteratively implementing and evaluating small refinements to the project so they are aware of their
progress. At an affective level, the educator should be aware that the students may become
discouraged if their selected learning method fails and be on hand to discuss alternatives, while
acknowledging the hard work they have already committed to the task.
In order to support the development of effective self evaluation skills, the educator has a role in
structuring the learning environment with tasks which develop peer and self assessment skills, but
also by modeling the necessary reflective skills. Public evaluation of specific features of good work
can assist students in understanding standards, as well as providing positive feedback for learners.
There may occasionally be students who find the success of other students demoralizing and who
are discouraged from trying because they feel they can never match the performance of their peers.
This can be particularly problematic in environments where able students “strut” by demonstrating
their prowess at the expense of others (Garvin-Doxas and Barker, 2004). Therefore, care should be
taken to acknowledge the effort and improvements in learning as much as the actual product itself.
Anonymous examples of common mistakes (perhaps drawn from previous cohorts) can also be
39
beneficial in enabling students to identify and correct similar problems in their own work. There
are also a number of issues to be aware of relating to the affective component of self and peer
evaluation. The educator can explicitly discuss helpful attitudes towards evaluation to help less
confident students find it less intimidating. These attitudes include the belief that performance can
be improved through practice; that students with little prior knowledge can make progress as well
as those with existing experience; that it is acceptable to be proud of one’s own good work; that it
might be necessary to be patient with one’s progress initially as skills develop; and that it is more
productive to plan how to recover from poor progress and failure than blaming oneself for lack of
success.
Inexperienced learners may also need help in interpreting peer review. The educator should model
constructive peer feedback to minimize the chance of students becoming demoralized after a
negative peer review. However, the students should also be encouraged to go beyond simply
praising their peer’s work and offer specific suggestions for improvement. After receiving peer
suggestions, the student may require some help in deciding which of the suggestions are possible to
adopt in the time remaining, and indeed whether the suggestions would benefit the project. The
latter is particularly important in creative projects where there may simply be a difference of
aesthetic values between reviewer and reviewee.
Towards the middle of the project, as the learners have developed the requisite technical skills to
start independent work on their projects, it would be beneficial to discuss productive working
practices with the class. They should be reminded to evaluate their current progress against their
expected progress to help them set more realistic targets next time. Similarly, it is helpful for them
to evaluate the difficulty of their chosen goal once they have attempted it so that they can modify
future goals to be more or less ambitious. As some students may not be engaging with the project at
this stage due to commitments to other coursework, educators may find it useful to have a way of
identifying “at risk” students whose projects are floundering. Possible methods for doing this
including setting up an automatic alert in the VLE to notify instructors when students have not
posted to their blog for a specified period of time, and checking this against class attendance
records for students, or discussing project progress with students during lab classes.
Setting up a positive learning environment is important from the outset. The educator can explicitly
encourage peer support through structured opportunities such as pair exercises, peer review and
blog commenting. The students should be aware that it is acceptable to seek help from staff and
peers, but should be clear on the boundaries of what constitutes plagiarism. With big classes, it is
worthwhile to explicitly discuss expectations about what level of help staff realistically have time to
provide, and alternative sources of help students can use if no staff member is available. Lastly,
particularly in technical subjects, students may be critical of the constraints of the learning
environment or the resources available, and while it is useful for them to develop the analytical
skills of identifying these flaws, the educator may need to be prepared to explain why the
constraints exist and negotiate solutions.
Naturally there are some caveats about these recommendations. Not all effective self directed
learners display all of the skills identified in Table 15, and nor do all less effective learners suffer
from all of these difficulties. Nonetheless, these markers may be a guide for other educators when
40
diagnosing their students’ self directed learning skills. There are likely to be gaps in the skills or
difficulties identified here as they are based on evidence from one particular class in the computer
science domain. Educators working in different subjects or at a different educational level may have
different experiences. In spite of these limitations, however, it is hoped that the recommendations
will be useful in providing a starting point for other educators who are considering how to
structure the learning environment to facilitate self directed learning.
41
Skill Sub-skill More effective learners…
Less effective learners …
Advice for educators
Sample prompt on lab sheet
When
Planning Generating own goals
Playfully explore the possibilities offered by the learning environment. Devise mini-projects to learn skills. Extend lab exercises. Set own challenges if module is too easy.
Don’t set their own goals in the early stages of learning a new environment. Stick to tasks and materials from staff. Note the module is too easy but don’t take responsibility for setting themselves harder tasks.
Explain to learners that learning through exploration can be valuable. Encourage able learners who seem bored to challenge themselves.
Did you discover anything interesting while exploring the software today? Can you think of a small project to carry out in the next lab session? Finding this too easy? Write down a task which would challenge you more and attempt it at home.
Early on in exploratory stages of creative project
Planning Generating and selecting ideas
Try to set the challenge of an idea according to their abilities. Consider multiple ideas. Analyse the difficulty of the idea in some detail before committing. Defer decision making until they can estimate their abilities more accurately. Are flexible about adapting or scaling back ideas if necessary. Can judge when to stop pursuing a
Lack self efficacy. Don’t investigate the feasibility of ideas before selection. Overestimate their ability and choose an idea which is too hard. Underestimate their ability and choose an idea which is too easy.
Discuss plans with students to advise on how difficult an idea will be to implement. Divert students from dead ends before they invest too much effort. Encourage less confident but able students to be more ambitious. Encourage students to think of several alternatives. Help students deal with
Now try brain storming some other ideas for your project. Which of the ideas do you think would be easiest to do? Which would be hardest? Which would match your current skill level best? (Note: sometimes it is a good idea to defer making a decision until you have had
At the stage when students are beginning to make decisions about which creative project to work on
42
dead end. frustration. Celebrate creative ideas.
more practice)
Planning Planning learning
Plan what to work on next after each session. Make lists of items to research or problems to fix. Plan learning strategy. Are able to re-plan and choose another method if the original method fails. Decide to take risks. Decide to aim high. Decide to invest effort. Consider development process as a whole.
Decide to minimise effort. Get stuck if the first method they try fails.
Encourage learners to finish each session with a plan for next steps. Discuss with students alternative methods they could use if one fails. Reward effort.
What will you work on in your next work session? What problems do you need to fix for next time? What is your plan for carrying out the project? If things are not working well for you: can you think of another method of solving this problem which might work better?
After each class
Planning Task management
Split tasks into subtasks with manageable deadlines. Set themselves deadlines from early in the project. Work iteratively on multiple “drafts” – increase standards at each iteration. Recognise when a general framework could help subparts of the solution.
Only use external deadlines rather than setting their own interim deadlines. Make vague plans, or none at all. Make unrealistic plans.
Encourage learners to set deadlines for themselves throughout the module. Help them to identify task dependencies. Encourage them to consider using an iterative development methodology.
What small tasks will you divide your project into? Which order should you do the tasks in? For each of the small tasks, when do you aim to complete them?
Every week
43
Recognise task dependencies.
Evaluating Evaluating own performance
Evaluate their progress regularly. Look back on progress over a few weeks. Recognise that skills require effort and practice to develop. Analyse in some detail which skills they have mastered and which need more work. Consider possible reasons for perceived poor progress. Evaluate whether their current strategy is working. Assess their level of understanding and what activities increase understanding. Can reflect honestly on what they have learned during a module and how they could have done better.
Worry about being stupid. Lack metacognitive knowledge about the task domain (such as about the nature of problem solving). Judge themselves harshly even if external evidence suggests they are doing well. Do not reflect in detail on their learning – can be vague.
Scaffold evaluation skills by self and peer assessment exercises. Encourage and model regular reflection. Advise students to focus on solutions to lack of progress rather than dwelling on failures. Persuade students that it is acceptable to acknowledge when one has done well. Make it clear that there is a distinction between ability and experience; practice and effort will pay off even for students who start off with no prior knowledge. Remind learners that other people have difficulty with the material too. Reiterate that practice makes perfect.
How well did you work go today? Why? How well do you think you have been getting on in the module in the last couple of weeks? Why? If you are not happy with your progress, what do you think has been holding you back? How can you improve for the future? If you have been getting on well, which activities do you think have been most useful in helping you learn?
Micro level – every session Macro level – every couple of weeks
44
Evaluating Evaluating own product
Evaluate product so far critically in some detail. Evaluate engineering process as well as product. Decide whether an idea is good enough. Decide whether and how to take advice from a peer. If appropriate stand by own idea instead of taking peer advice.
Find it hard to judge the quality of their product. Are unduly critical of their own work.
Model evaluation of product by publically identifying good aspects of student work. Encourage learners to be patient with the quality of their product until their skills increase. Help learners to decide which aspects of peer advice are feasible within the time limit. Encourage learners to take pride in the quality of their engineering as well as the appearance or functionality. Reassure learners that it is acceptable to stick with one’s original ideas instead of taking peer review suggestions on board all of the time.
What are the strengths and weaknesses of your project? Do you agree with the advice offered during the peer review session? Are the suggestions offered in the peer review feasible?
Around ¾ of the way through a creative project
Evaluating Evaluating/ predicting task difficulty
Can accurately estimate how difficult a task will be. Can evaluate which parts of a task were most challenging after
Are pessimistic about task difficulty. Are optimistic about task difficulty. Assume that something will be too hard,
Suggest that learners compare expected to actual task difficulty to remind them that they are capable of mastering
Look back at your plans for last week. Did you achieve what you set out to do in the time you allowed yourself?
Weekly/ after a learning session
45
completing it. Regularly compare expectations of difficulty to actual difficulty. Recognise that effort is a predictor of task success.
and so don’t attempt it.
difficult material and also help them predict more realistically in future.
Was the last task you set yourself as difficult (or easy) as you thought it would be? Which part of the task was most difficult?
Evaluating Monitoring Know when to stop using a strategy if it isn’t working. Set time limits on exploring a solution to prevent dead ends. Recognise that some tasks are hard to estimate. Direct more time and effort into the project when it is going badly.
Drop standards rather than increase effort when progress is poor. Panic and work in a non-productive pattern (such as all night).
Discuss effective working practices with the class. Identify at-risk students and help them re-plan their project before it goes too badly wrong.
How do you feel your project is progressing? If you feel your project is in trouble, what can you do to put it back on track? Is it worth pursuing your current approach or should you try something else? Is there another strategy you could use to solve the problem?
Regularly from about half way through the project.
Evaluating Evaluating learning environment
Predict problems arising from constraints of learning environment, and plan around them. Evaluate challenges inherent to learning environment or activity. Compare learning environment
Blame the learning environment rather than their effort for lack of progress. Are critical of the learning environment without thinking through why it has particular constraints.
Acknowledge limitations with learning environment, but help the students find work arounds. Explain to learners why constraints or limitations exist.
Think about the learning environment you are working in (for example the technical set-up, the books and materials you use, the classes you attend and people you consult for help). What do you find
Around half way through the project, once the student has some experience with the learning environment
46
to previous learning environments. Learn how to work within constraints of learning environment.
most effective in this set-up? Is there anything which you find non-productive? If so, what could you do to change this?
Help seeking
From staff Ask for help before a problem becomes a major hindrance. Discuss ideas as well as problems with staff. Are aware there is satisfaction in solving one’s own problems.
Avoid asking staff for help because they worry abut being judged negatively, or because they fear becoming too dependent.
Reassure students that it is acceptable to ask for help when necessary. Advertise office hours or other time when staff are free to discuss projects.
If you are having difficulty, is there someone you could ask for help – the lecturer? Another student? If you have a good idea, why not discuss it with a member of staff or another student? Please be willing to give other students help and ideas if they ask.
Reminders in lectures or labs
Help seeking
From peers Work with other students to solve problems. Ask more able peers for help. Ask advice and feedback from peers. Give as well as receive peer support.
Use other students’ solutions without critical thought. Avoid working with peers Behave in a non-supportive way towards peers. Show off their knowledge in a way which makes other students feel inadequate.
Facilitate a supportive learning environment where students are expected to give and receive peer support. Model constructive feedback skills. Moderate “strutting” behaviour from more able students so less able students do not get demoralised.
Table 15. Hallmarks in blogging of effective and less effective self-directed learners
47
CONCLUSIONS The development of self directed learning skills is an important goal of higher education,
particularly in the first year. The experiences and qualitative analysis from a module for first and
second year computer science indicates that such skills can be fostered in a learning environment
with the following features. Firstly, students should be given an opportunity to direct their learning,
but in a supportive environment with approachable staff to guide them. Directing learning involves
not only regulating one’s performance and planning sub-tasks, but also selecting one’s own learning
goals in the first place. This can be done within a loose structure suggested by educators who
understand the skills and knowledge required in the domain. A flexible creative design project in
which the students could pick which product they wanted to make was effective in this study.
Secondly, the role of the educator is partly to provide advice on how to approach and manage a
project of this nature. In this capacity, the educator can model the processes the learner undergoes
throughout the lifecycle of a project, signpost which stages of the project students should typically
be tackling at any given time, demonstrate effective self assessment and peer review skills and
advise students on how to recover when the project goes wrong. Reflective writing in the form of
blogs can be beneficial here both as a diagnostic tool for educators to monitor students’ progress on
the domain task and in the development of their self directed learning skills. Further, the act of
producing a blog gives the students an opportunity for reflection which enables them to monitor
their own progress. In this learning situation at least, Williams and Jacob’s early optimistic view of
the educational potential of blogs was justified (Williams and Jacobs, 2004).
An extremely important aspect of the educator’s role, which should not be forgotten, is setting up a
positive affective learning environment in which students feel supported both by staff and their
peers. This involves monitoring the students’ affect with respect to their learning, bolstering
students with low self efficacy, managing expectations and helping them to deal with failure and
frustration. Learning to direct one’s own learning can initially be difficult and daunting; we owe it to
our students to support them throughout this journey.
ACKNOWLEDGEMENTS Many thanks to Nicole Kipar for her support and assistance throughout the project, and for second
rating the corpus. Roger Rist and Irene Higson’s contribution to teaching and marking was essential
to the success of the module. Dominic Seymour kindly wrote a program to extract the learning logs
from Blackboard and automatically word count them.
REFERENCES
48
ALEXANDER, P. 2008. Why This and Why Now? Introduction to the Special Issue on Metacognition, Self-Regulation, and Self-Regulated Learning. Educational Psychology Review, 20, 369-372.
ANDERSON, D., NASHON, S. & THOMAS, G. 2009. Evolution of Research Methods for Probing and Understanding Metacognition. Research in Science Education, 39, 181-195.
BOEKAERTS, M. 2003. Towards a model that integrates motivation, affect and learning. BJEP Monograph Series II, Number 2 - Development and Motivation, 1, 173-189.
BOUD, D. 2001. Using Journal Writing to Enhace Reflective Practice. NEW DIRECTIONS FOR ADULT AND CONTINUING EDUCATION,, 20, 9-17.
BROWN, A. 1987. Metacognition, executive control, selfregulation
and other more mysterious mechanisms. In: WEINERT, F. E. & KLUWE, R. H. (eds.) Metacognition, motivation and understanding. Hillsdale, New Jersey: Lawrence Erlbaum Associates.
COLLINS, A., BROWN, J. & HOLUM, A. 1991. Cognitive Apprenticeship: Making Things Visible. American Educator, 15, 6-11, 38-39.
COUTINHO, S. & NEUMAN, G. 2008. A model of metacognition, achievement goal orientation, learning style and self-efficacy. Learning Environments Research, 11, 131-151.
DAVIS, E. 2003. Prompting Middle School Science Students for Productive Reflection: Generic and Directed Prompts. Journal of the Learning Sciences, 12, 91-142.
FERDIG, R. E. & TRAMMELL, K. D. 2004. Content delivery in the 'Blogosphere' [Online]. Available: http://www.thejournal.com/magazine/vault/articleprintversion.cfm?aid=4677 [Accessed 19/4 2010].
FLAVELL, J. 1979. Metacognition and Cognitive Monitoring. American Psychologist, 34, 906-911. FLEISS, J. L. 1981. Statistical methods for rates and proportions, New York ; Chichester, Wiley. GARVIN-DOXAS, K. & BARKER, L. J. 2004. Communication in computer science classrooms:
understanding defensive climates as a means of creating supportive behaviors. Journal of Educational. Resources in Computing, 4.
HAYES, N. 2000. Doing psychological research : gathering and analysing data, Buckingham, Open University Press.
HIGGINS, S. 2009. Learning to Learn. Beyond Current Horizons. Durham: Beyond Current Horizons Project.
KAPLAN, A. 2008. Clarifying Metacognition, Self-Regulation, and Self-Regulated Learning: What’s the Purpose? Educational Psychology Review, 20, 477-484.
KERAWALLA, L., MINOCHA, S., KIRKUP, G. & CONOLE, G. 2009. An empirically grounded framework to guide blogging in higher education. Journal of Computer Assisted Learning, 25, 31-42.
LOH, B., RADINSKY, J., RUSSELL, E., GOMEZ, L., REISER, B. & EDELSON, D. 1998. The progress portfolio: designing reflective tools for a classroom context. Proceedings of the SIGCHI conference on Human factors in computing systems. Los Angeles, California, United States: ACM Press/Addison-Wesley Publishing Co.
LOYENS, S., MAGDA, J. & RIKERS, R. 2008. Self-Directed Learning in Problem-Based Learning and its Relationships with Self-Regulated Learning. Educational Psychology Review, 20, 411-427.
LUEHMANN, A. 2008. Using Blogging in Support of Teacher Professional Identity Development: A Case Study. Journal of the Learning Sciences, 17, p287-337.
MAYES, T. 2009. Overview of the enhancement theme 2006-08: The aims, themes and challenges. Mansfield: The Quality Assurance Agency for Higher Education.
MINOCHA, S. 2009a. An empirically-grounded study on the effective use of social software in education. Education + Training, 51, 381 - 394.
MINOCHA, S. 2009b. Role of social software tools in education: a literature review. Education + Training, 51, 353-369.
MOORES, T., CHANG, J. C.-J. & SMITH, D., K. 2006. Clarifying the role of self-efficacy and metacognition as predictors of performance: construct development and test. SIGMIS Database, 37, 125-132.
49
NOUSHAD, P. 2008. Cognitions about Cognitions: The Theory of Metacognition. SCARDAMALIA, M. & BEREITER, C. 1991. Higher Levels of Agency in Knowledge Building: A
Challenge for the Design of New Knowledge Media. Journal of the Learning Sciences, 1, 37-68.
STONE, J. A. & MADIGAN, E. M. 2007. Integrating reflective writing in CS/IS. SIGCSE Bull., 39, 42-45. VRUGT, A. & OORT, F. 2008. Metacognition, achievement goals, study strategies and academic
achievement: pathways to achievement. Metacognition and Learning, 3, 123-146. WHITE, B. & FREDERIKSEN, J. 1998. Inquiry, Modeling, and Metacognition: Making Science
Acessible to All Students. Cognition and Instruction, 16, 3-118. WILLIAMS, J. & JACOBS, J. 2004. Exploring the use of blogs as learning spaces in the higher
education sector. Australasian Journal of Educational Technology, 20, 232-247.